diff --git "a/train.log" "b/train.log" new file mode 100644--- /dev/null +++ "b/train.log" @@ -0,0 +1,10014 @@ +2026-08-25 03:51:06 | INFO | train rank=0 | Starting training project=codebase_v7_droid_steps100k_bs128_lr1e-5_qwen3vl2b_new_policy save_dir=/data/NTU_slab/draven/checkpoints/codebase/codebase_v7_droid/steps100k_bs128_lr1e-5_qwen3vl2b_new_policy distributed=True deepspeed=True world_size=8 device=cuda:0 +2026-08-25 03:51:31 | INFO | train rank=0 | No pretrained checkpoint provided; training from scratch +2026-08-25 03:51:31 | INFO | train rank=0 | Initializing Droid LeRobot Dataset: path=/data/NTU_slab/draven/data/DROID fps=15.00 +2026-08-25 03:51:31 | INFO | train rank=0 | Batch config: total_batch_size=128 world_size=8 grad_acc_steps=1 per_device_batch_size=16 +2026-08-25 03:51:31 | INFO | train rank=0 | Dataset image_resize_size=256 +2026-08-25 03:51:31 | INFO | train rank=0 | Droid uses parquet-shard streaming with a per-rank shuffle buffer: batch_size=16 buffer_size=1024 num_workers=8 drop_last=False +2026-08-25 03:51:39 | INFO | train rank=0 | Droid streaming layout: parquet_shards=50 active_workers=8 worker_buffer_min=128 worker_buffer_max=128 max_active_shards=2 +2026-08-25 03:51:39 | INFO | train rank=0 | Resolved Droid action stats from /data/NTU_slab/draven/data/DROID/meta/stats.json: min=[-0.2077193409204483, -0.8572492003440857, -0.28132379055023193, -3.141592502593994, -1.570475697517395, -3.1415903568267822] max=[0.9295652508735657, 0.8648782968521118, 1.074978232383728, 3.1415927410125732, 1.569351077079773, 3.1415891647338867] +2026-08-25 03:51:39 | INFO | train rank=0 | Trainable params: 2,952,988,679 / 2,952,988,679 +2026-08-25 03:55:30 | INFO | train rank=0 | Starting training start_step=0 max_steps=100000 remaining_steps=100000 log_interval=10 save_interval=100000 +2026-08-25 03:56:58 | INFO | train rank=0 | step=10/100000 progress=0.0% loss=1.273438 lr=1.800000e-08 elapsed=00:01:28 avg_step=8.80s recent_step=8.80s eta=244:30:58 +2026-08-25 03:57:11 | INFO | train rank=0 | step=20/100000 progress=0.0% loss=1.281250 lr=3.800000e-08 elapsed=00:01:40 avg_step=5.05s recent_step=1.29s eta=140:09:07 +2026-08-25 03:57:24 | INFO | train rank=0 | step=30/100000 progress=0.0% loss=1.320312 lr=5.800000e-08 elapsed=00:01:54 avg_step=3.82s recent_step=1.38s eta=106:10:44 +2026-08-25 03:57:36 | INFO | train rank=0 | step=40/100000 progress=0.0% loss=1.246094 lr=7.800000e-08 elapsed=00:02:06 avg_step=3.16s recent_step=1.15s eta=87:38:13 +2026-08-25 03:57:50 | INFO | train rank=0 | step=50/100000 progress=0.1% loss=1.275391 lr=9.800000e-08 elapsed=00:02:20 avg_step=2.81s recent_step=1.43s eta=78:03:01 +2026-08-25 03:58:05 | INFO | train rank=0 | step=60/100000 progress=0.1% loss=1.311523 lr=1.180000e-07 elapsed=00:02:35 avg_step=2.60s recent_step=1.52s eta=72:04:36 +2026-08-25 03:58:17 | INFO | train rank=0 | step=70/100000 progress=0.1% loss=1.286133 lr=1.380000e-07 elapsed=00:02:47 avg_step=2.39s recent_step=1.14s eta=66:17:36 +2026-08-25 03:58:29 | INFO | train rank=0 | step=80/100000 progress=0.1% loss=1.257812 lr=1.580000e-07 elapsed=00:02:59 avg_step=2.25s recent_step=1.26s eta=62:21:37 +2026-08-25 03:58:41 | INFO | train rank=0 | step=90/100000 progress=0.1% loss=1.310547 lr=1.780000e-07 elapsed=00:03:11 avg_step=2.13s recent_step=1.19s eta=59:05:10 +2026-08-25 03:58:55 | INFO | train rank=0 | step=100/100000 progress=0.1% loss=1.309570 lr=1.980000e-07 elapsed=00:03:25 avg_step=2.05s recent_step=1.35s eta=56:55:02 +2026-08-25 03:59:09 | INFO | train rank=0 | step=110/100000 progress=0.1% loss=1.264648 lr=2.180000e-07 elapsed=00:03:39 avg_step=1.99s recent_step=1.41s eta=55:18:25 +2026-08-25 03:59:25 | INFO | train rank=0 | step=120/100000 progress=0.1% loss=1.221680 lr=2.380000e-07 elapsed=00:03:55 avg_step=1.96s recent_step=1.65s eta=54:29:54 +2026-08-25 03:59:38 | INFO | train rank=0 | step=130/100000 progress=0.1% loss=1.240234 lr=2.580000e-07 elapsed=00:04:08 avg_step=1.91s recent_step=1.25s eta=52:58:28 +2026-08-25 03:59:53 | INFO | train rank=0 | step=140/100000 progress=0.1% loss=1.290039 lr=2.780000e-07 elapsed=00:04:23 avg_step=1.88s recent_step=1.52s eta=52:11:21 +2026-08-25 04:00:04 | INFO | train rank=0 | step=150/100000 progress=0.1% loss=1.336914 lr=2.980000e-07 elapsed=00:04:34 avg_step=1.83s recent_step=1.14s eta=50:48:21 +2026-08-25 04:00:17 | INFO | train rank=0 | step=160/100000 progress=0.2% loss=1.361328 lr=3.180000e-07 elapsed=00:04:47 avg_step=1.80s recent_step=1.25s eta=49:47:24 +2026-08-25 04:00:30 | INFO | train rank=0 | step=170/100000 progress=0.2% loss=1.281250 lr=3.380000e-07 elapsed=00:05:00 avg_step=1.77s recent_step=1.35s eta=49:03:27 +2026-08-25 04:00:42 | INFO | train rank=0 | step=180/100000 progress=0.2% loss=1.231445 lr=3.580000e-07 elapsed=00:05:12 avg_step=1.74s recent_step=1.17s eta=48:07:31 +2026-08-25 04:00:54 | INFO | train rank=0 | step=190/100000 progress=0.2% loss=1.275391 lr=3.780000e-07 elapsed=00:05:24 avg_step=1.71s recent_step=1.24s eta=47:23:40 +2026-08-25 04:01:05 | INFO | train rank=0 | step=200/100000 progress=0.2% loss=1.218750 lr=3.980000e-07 elapsed=00:05:35 avg_step=1.68s recent_step=1.11s eta=46:33:08 +2026-08-25 04:01:18 | INFO | train rank=0 | step=210/100000 progress=0.2% loss=1.304688 lr=4.180000e-07 elapsed=00:05:48 avg_step=1.66s recent_step=1.22s eta=45:56:41 +2026-08-25 04:01:30 | INFO | train rank=0 | step=220/100000 progress=0.2% loss=1.270508 lr=4.380000e-07 elapsed=00:06:00 avg_step=1.64s recent_step=1.23s eta=45:24:16 +2026-08-25 04:01:44 | INFO | train rank=0 | step=230/100000 progress=0.2% loss=1.282227 lr=4.580000e-07 elapsed=00:06:14 avg_step=1.63s recent_step=1.38s eta=45:05:00 +2026-08-25 04:01:56 | INFO | train rank=0 | step=240/100000 progress=0.2% loss=1.300781 lr=4.780000e-07 elapsed=00:06:26 avg_step=1.61s recent_step=1.24s eta=44:37:49 +2026-08-25 04:02:08 | INFO | train rank=0 | step=250/100000 progress=0.2% loss=1.249023 lr=4.980000e-07 elapsed=00:06:37 avg_step=1.59s recent_step=1.14s eta=44:06:31 +2026-08-25 04:02:23 | INFO | train rank=0 | step=260/100000 progress=0.3% loss=1.264648 lr=5.180000e-07 elapsed=00:06:53 avg_step=1.59s recent_step=1.52s eta=44:01:53 +2026-08-25 04:02:35 | INFO | train rank=0 | step=270/100000 progress=0.3% loss=1.284180 lr=5.380000e-07 elapsed=00:07:05 avg_step=1.58s recent_step=1.27s eta=43:41:49 +2026-08-25 04:02:49 | INFO | train rank=0 | step=280/100000 progress=0.3% loss=1.218750 lr=5.580000e-07 elapsed=00:07:19 avg_step=1.57s recent_step=1.40s eta=43:30:48 +2026-08-25 04:03:02 | INFO | train rank=0 | step=290/100000 progress=0.3% loss=1.266602 lr=5.780000e-07 elapsed=00:07:32 avg_step=1.56s recent_step=1.26s eta=43:12:49 +2026-08-25 04:03:15 | INFO | train rank=0 | step=300/100000 progress=0.3% loss=1.285156 lr=5.980000e-07 elapsed=00:07:44 avg_step=1.55s recent_step=1.25s eta=42:55:09 +2026-08-25 04:03:28 | INFO | train rank=0 | step=310/100000 progress=0.3% loss=1.315430 lr=6.180000e-07 elapsed=00:07:57 avg_step=1.54s recent_step=1.31s eta=42:41:52 +2026-08-25 04:03:41 | INFO | train rank=0 | step=320/100000 progress=0.3% loss=1.250977 lr=6.380000e-07 elapsed=00:08:11 avg_step=1.54s recent_step=1.36s eta=42:32:06 +2026-08-25 04:03:54 | INFO | train rank=0 | step=330/100000 progress=0.3% loss=1.266602 lr=6.580000e-07 elapsed=00:08:24 avg_step=1.53s recent_step=1.27s eta=42:18:30 +2026-08-25 04:04:09 | INFO | train rank=0 | step=340/100000 progress=0.3% loss=1.296875 lr=6.780000e-07 elapsed=00:08:39 avg_step=1.53s recent_step=1.51s eta=42:17:34 +2026-08-25 04:04:21 | INFO | train rank=0 | step=350/100000 progress=0.4% loss=1.193359 lr=6.980000e-07 elapsed=00:08:51 avg_step=1.52s recent_step=1.19s eta=42:01:28 +2026-08-25 04:04:32 | INFO | train rank=0 | step=360/100000 progress=0.4% loss=1.302734 lr=7.180000e-07 elapsed=00:09:02 avg_step=1.51s recent_step=1.09s eta=41:41:37 +2026-08-25 04:04:45 | INFO | train rank=0 | step=370/100000 progress=0.4% loss=1.267578 lr=7.380000e-07 elapsed=00:09:15 avg_step=1.50s recent_step=1.28s eta=41:31:06 +2026-08-25 04:05:02 | INFO | train rank=0 | step=380/100000 progress=0.4% loss=1.250977 lr=7.580000e-07 elapsed=00:09:32 avg_step=1.51s recent_step=1.75s eta=41:41:37 +2026-08-25 04:05:14 | INFO | train rank=0 | step=390/100000 progress=0.4% loss=1.252930 lr=7.780000e-07 elapsed=00:09:44 avg_step=1.50s recent_step=1.16s eta=41:26:28 +2026-08-25 04:05:28 | INFO | train rank=0 | step=400/100000 progress=0.4% loss=1.318359 lr=7.980000e-07 elapsed=00:09:58 avg_step=1.50s recent_step=1.44s eta=41:23:38 +2026-08-25 04:05:39 | INFO | train rank=0 | step=410/100000 progress=0.4% loss=1.262695 lr=8.180000e-07 elapsed=00:10:09 avg_step=1.49s recent_step=1.08s eta=41:06:31 +2026-08-25 04:05:52 | INFO | train rank=0 | step=420/100000 progress=0.4% loss=1.337891 lr=8.380000e-07 elapsed=00:10:22 avg_step=1.48s recent_step=1.31s eta=40:59:27 +2026-08-25 04:06:05 | INFO | train rank=0 | step=430/100000 progress=0.4% loss=1.299805 lr=8.580000e-07 elapsed=00:10:35 avg_step=1.48s recent_step=1.29s eta=40:51:38 +2026-08-25 04:06:19 | INFO | train rank=0 | step=440/100000 progress=0.4% loss=1.309570 lr=8.780000e-07 elapsed=00:10:49 avg_step=1.48s recent_step=1.42s eta=40:49:07 +2026-08-25 04:06:34 | INFO | train rank=0 | step=450/100000 progress=0.4% loss=1.296875 lr=8.980000e-07 elapsed=00:11:04 avg_step=1.48s recent_step=1.50s eta=40:49:56 +2026-08-25 04:06:48 | INFO | train rank=0 | step=460/100000 progress=0.5% loss=1.311523 lr=9.180000e-07 elapsed=00:11:18 avg_step=1.48s recent_step=1.43s eta=40:47:52 +2026-08-25 04:07:03 | INFO | train rank=0 | step=470/100000 progress=0.5% loss=1.227539 lr=9.380000e-07 elapsed=00:11:33 avg_step=1.47s recent_step=1.44s eta=40:46:15 +2026-08-25 04:07:15 | INFO | train rank=0 | step=480/100000 progress=0.5% loss=1.255859 lr=9.580000e-07 elapsed=00:11:45 avg_step=1.47s recent_step=1.19s eta=40:36:13 +2026-08-25 04:07:26 | INFO | train rank=0 | step=490/100000 progress=0.5% loss=1.272461 lr=9.780000e-07 elapsed=00:11:56 avg_step=1.46s recent_step=1.13s eta=40:24:27 +2026-08-25 04:07:38 | INFO | train rank=0 | step=500/100000 progress=0.5% loss=1.250000 lr=9.980000e-07 elapsed=00:12:08 avg_step=1.46s recent_step=1.20s eta=40:15:33 +2026-08-25 04:07:53 | INFO | train rank=0 | step=510/100000 progress=0.5% loss=1.253906 lr=1.018000e-06 elapsed=00:12:23 avg_step=1.46s recent_step=1.51s eta=40:17:03 +2026-08-25 04:08:04 | INFO | train rank=0 | step=520/100000 progress=0.5% loss=1.187500 lr=1.038000e-06 elapsed=00:12:34 avg_step=1.45s recent_step=1.14s eta=40:06:31 +2026-08-25 04:08:17 | INFO | train rank=0 | step=530/100000 progress=0.5% loss=1.177734 lr=1.058000e-06 elapsed=00:12:47 avg_step=1.45s recent_step=1.25s eta=39:59:59 +2026-08-25 04:08:30 | INFO | train rank=0 | step=540/100000 progress=0.5% loss=1.218750 lr=1.078000e-06 elapsed=00:13:00 avg_step=1.44s recent_step=1.30s eta=39:55:15 +2026-08-25 04:08:41 | INFO | train rank=0 | step=550/100000 progress=0.5% loss=1.203125 lr=1.098000e-06 elapsed=00:13:11 avg_step=1.44s recent_step=1.16s eta=39:46:20 +2026-08-25 04:08:54 | INFO | train rank=0 | step=560/100000 progress=0.6% loss=1.211914 lr=1.118000e-06 elapsed=00:13:24 avg_step=1.44s recent_step=1.27s eta=39:40:57 +2026-08-25 04:09:05 | INFO | train rank=0 | step=570/100000 progress=0.6% loss=1.190430 lr=1.138000e-06 elapsed=00:13:35 avg_step=1.43s recent_step=1.13s eta=39:31:52 +2026-08-25 04:09:19 | INFO | train rank=0 | step=580/100000 progress=0.6% loss=1.095703 lr=1.158000e-06 elapsed=00:13:49 avg_step=1.43s recent_step=1.39s eta=39:30:28 +2026-08-25 04:09:35 | INFO | train rank=0 | step=590/100000 progress=0.6% loss=1.180664 lr=1.178000e-06 elapsed=00:14:05 avg_step=1.43s recent_step=1.55s eta=39:33:32 +2026-08-25 04:09:46 | INFO | train rank=0 | step=600/100000 progress=0.6% loss=1.282227 lr=1.198000e-06 elapsed=00:14:16 avg_step=1.43s recent_step=1.13s eta=39:24:52 +2026-08-25 04:09:58 | INFO | train rank=0 | step=610/100000 progress=0.6% loss=1.171875 lr=1.218000e-06 elapsed=00:14:28 avg_step=1.42s recent_step=1.19s eta=39:18:15 +2026-08-25 04:10:14 | INFO | train rank=0 | step=620/100000 progress=0.6% loss=1.121094 lr=1.238000e-06 elapsed=00:14:44 avg_step=1.43s recent_step=1.62s eta=39:23:08 +2026-08-25 04:10:26 | INFO | train rank=0 | step=630/100000 progress=0.6% loss=1.180664 lr=1.258000e-06 elapsed=00:14:56 avg_step=1.42s recent_step=1.17s eta=39:16:03 +2026-08-25 04:10:38 | INFO | train rank=0 | step=640/100000 progress=0.6% loss=1.222656 lr=1.278000e-06 elapsed=00:15:08 avg_step=1.42s recent_step=1.25s eta=39:11:24 +2026-08-25 04:10:50 | INFO | train rank=0 | step=650/100000 progress=0.7% loss=1.215820 lr=1.298000e-06 elapsed=00:15:19 avg_step=1.42s recent_step=1.12s eta=39:03:25 +2026-08-25 04:11:02 | INFO | train rank=0 | step=660/100000 progress=0.7% loss=1.129883 lr=1.318000e-06 elapsed=00:15:32 avg_step=1.41s recent_step=1.24s eta=38:58:45 +2026-08-25 04:11:14 | INFO | train rank=0 | step=670/100000 progress=0.7% loss=1.149414 lr=1.338000e-06 elapsed=00:15:44 avg_step=1.41s recent_step=1.23s eta=38:53:58 +2026-08-25 04:11:26 | INFO | train rank=0 | step=680/100000 progress=0.7% loss=1.224609 lr=1.358000e-06 elapsed=00:15:56 avg_step=1.41s recent_step=1.19s eta=38:48:24 +2026-08-25 04:11:39 | INFO | train rank=0 | step=690/100000 progress=0.7% loss=1.166016 lr=1.378000e-06 elapsed=00:16:09 avg_step=1.40s recent_step=1.29s eta=38:45:27 +2026-08-25 04:11:51 | INFO | train rank=0 | step=700/100000 progress=0.7% loss=1.074219 lr=1.398000e-06 elapsed=00:16:21 avg_step=1.40s recent_step=1.18s eta=38:39:55 +2026-08-25 04:12:04 | INFO | train rank=0 | step=710/100000 progress=0.7% loss=1.112305 lr=1.418000e-06 elapsed=00:16:34 avg_step=1.40s recent_step=1.30s eta=38:37:25 +2026-08-25 04:12:17 | INFO | train rank=0 | step=720/100000 progress=0.7% loss=1.126953 lr=1.438000e-06 elapsed=00:16:46 avg_step=1.40s recent_step=1.26s eta=38:34:01 +2026-08-25 04:12:30 | INFO | train rank=0 | step=730/100000 progress=0.7% loss=1.072266 lr=1.458000e-06 elapsed=00:16:59 avg_step=1.40s recent_step=1.30s eta=38:31:33 +2026-08-25 04:12:42 | INFO | train rank=0 | step=740/100000 progress=0.7% loss=1.077637 lr=1.478000e-06 elapsed=00:17:11 avg_step=1.39s recent_step=1.21s eta=38:27:03 +2026-08-25 04:12:55 | INFO | train rank=0 | step=750/100000 progress=0.8% loss=1.201172 lr=1.498000e-06 elapsed=00:17:25 avg_step=1.39s recent_step=1.38s eta=38:26:35 +2026-08-25 04:13:07 | INFO | train rank=0 | step=760/100000 progress=0.8% loss=1.156250 lr=1.518000e-06 elapsed=00:17:37 avg_step=1.39s recent_step=1.19s eta=38:21:54 +2026-08-25 04:13:22 | INFO | train rank=0 | step=770/100000 progress=0.8% loss=1.056641 lr=1.538000e-06 elapsed=00:17:52 avg_step=1.39s recent_step=1.48s eta=38:23:32 +2026-08-25 04:13:34 | INFO | train rank=0 | step=780/100000 progress=0.8% loss=1.117188 lr=1.558000e-06 elapsed=00:18:04 avg_step=1.39s recent_step=1.23s eta=38:19:50 +2026-08-25 04:13:46 | INFO | train rank=0 | step=790/100000 progress=0.8% loss=1.103516 lr=1.578000e-06 elapsed=00:18:16 avg_step=1.39s recent_step=1.14s eta=38:14:27 +2026-08-25 04:13:58 | INFO | train rank=0 | step=800/100000 progress=0.8% loss=1.063477 lr=1.598000e-06 elapsed=00:18:28 avg_step=1.39s recent_step=1.18s eta=38:09:56 +2026-08-25 04:14:11 | INFO | train rank=0 | step=810/100000 progress=0.8% loss=1.091797 lr=1.618000e-06 elapsed=00:18:41 avg_step=1.38s recent_step=1.35s eta=38:08:57 +2026-08-25 04:14:27 | INFO | train rank=0 | step=820/100000 progress=0.8% loss=1.103516 lr=1.638000e-06 elapsed=00:18:57 avg_step=1.39s recent_step=1.60s eta=38:13:07 +2026-08-25 04:14:39 | INFO | train rank=0 | step=830/100000 progress=0.8% loss=0.948242 lr=1.658000e-06 elapsed=00:19:09 avg_step=1.38s recent_step=1.16s eta=38:08:27 +2026-08-25 04:14:51 | INFO | train rank=0 | step=840/100000 progress=0.8% loss=1.024414 lr=1.678000e-06 elapsed=00:19:21 avg_step=1.38s recent_step=1.22s eta=38:05:01 +2026-08-25 04:15:03 | INFO | train rank=0 | step=850/100000 progress=0.9% loss=1.012695 lr=1.698000e-06 elapsed=00:19:33 avg_step=1.38s recent_step=1.21s eta=38:01:26 +2026-08-25 04:15:16 | INFO | train rank=0 | step=860/100000 progress=0.9% loss=0.955078 lr=1.718000e-06 elapsed=00:19:46 avg_step=1.38s recent_step=1.30s eta=37:59:36 +2026-08-25 04:15:28 | INFO | train rank=0 | step=870/100000 progress=0.9% loss=0.995117 lr=1.738000e-06 elapsed=00:19:58 avg_step=1.38s recent_step=1.23s eta=37:56:34 +2026-08-25 04:15:42 | INFO | train rank=0 | step=880/100000 progress=0.9% loss=0.909668 lr=1.758000e-06 elapsed=00:20:12 avg_step=1.38s recent_step=1.33s eta=37:55:22 +2026-08-25 04:15:54 | INFO | train rank=0 | step=890/100000 progress=0.9% loss=0.977051 lr=1.778000e-06 elapsed=00:20:24 avg_step=1.38s recent_step=1.22s eta=37:52:14 +2026-08-25 04:16:06 | INFO | train rank=0 | step=900/100000 progress=0.9% loss=0.947754 lr=1.798000e-06 elapsed=00:20:36 avg_step=1.37s recent_step=1.22s eta=37:49:05 +2026-08-25 04:16:18 | INFO | train rank=0 | step=910/100000 progress=0.9% loss=1.022949 lr=1.818000e-06 elapsed=00:20:48 avg_step=1.37s recent_step=1.20s eta=37:45:44 +2026-08-25 04:16:30 | INFO | train rank=0 | step=920/100000 progress=0.9% loss=0.898926 lr=1.838000e-06 elapsed=00:21:00 avg_step=1.37s recent_step=1.17s eta=37:41:54 +2026-08-25 04:16:42 | INFO | train rank=0 | step=930/100000 progress=0.9% loss=0.987305 lr=1.858000e-06 elapsed=00:21:12 avg_step=1.37s recent_step=1.22s eta=37:39:05 +2026-08-25 04:16:55 | INFO | train rank=0 | step=940/100000 progress=0.9% loss=0.953125 lr=1.878000e-06 elapsed=00:21:25 avg_step=1.37s recent_step=1.31s eta=37:37:55 +2026-08-25 04:17:08 | INFO | train rank=0 | step=950/100000 progress=0.9% loss=0.877441 lr=1.898000e-06 elapsed=00:21:38 avg_step=1.37s recent_step=1.30s eta=37:36:26 +2026-08-25 04:17:22 | INFO | train rank=0 | step=960/100000 progress=1.0% loss=0.860840 lr=1.918000e-06 elapsed=00:21:52 avg_step=1.37s recent_step=1.40s eta=37:36:47 +2026-08-25 04:17:33 | INFO | train rank=0 | step=970/100000 progress=1.0% loss=0.879883 lr=1.938000e-06 elapsed=00:22:03 avg_step=1.36s recent_step=1.14s eta=37:32:37 +2026-08-25 04:17:46 | INFO | train rank=0 | step=980/100000 progress=1.0% loss=0.898438 lr=1.958000e-06 elapsed=00:22:16 avg_step=1.36s recent_step=1.30s eta=37:31:14 +2026-08-25 04:18:00 | INFO | train rank=0 | step=990/100000 progress=1.0% loss=0.804688 lr=1.978000e-06 elapsed=00:22:30 avg_step=1.36s recent_step=1.38s eta=37:31:14 +2026-08-25 04:18:12 | INFO | train rank=0 | step=1000/100000 progress=1.0% loss=0.842773 lr=1.998000e-06 elapsed=00:22:42 avg_step=1.36s recent_step=1.17s eta=37:27:49 +2026-08-25 04:18:25 | INFO | train rank=0 | step=1010/100000 progress=1.0% loss=0.818359 lr=2.018000e-06 elapsed=00:22:55 avg_step=1.36s recent_step=1.32s eta=37:26:54 +2026-08-25 04:18:37 | INFO | train rank=0 | step=1020/100000 progress=1.0% loss=0.840332 lr=2.038000e-06 elapsed=00:23:07 avg_step=1.36s recent_step=1.18s eta=37:23:43 +2026-08-25 04:18:49 | INFO | train rank=0 | step=1030/100000 progress=1.0% loss=0.804688 lr=2.058000e-06 elapsed=00:23:18 avg_step=1.36s recent_step=1.16s eta=37:20:18 +2026-08-25 04:19:02 | INFO | train rank=0 | step=1040/100000 progress=1.0% loss=0.802734 lr=2.078000e-06 elapsed=00:23:32 avg_step=1.36s recent_step=1.37s eta=37:20:12 +2026-08-25 04:19:17 | INFO | train rank=0 | step=1050/100000 progress=1.1% loss=0.914062 lr=2.098000e-06 elapsed=00:23:47 avg_step=1.36s recent_step=1.50s eta=37:22:11 +2026-08-25 04:19:30 | INFO | train rank=0 | step=1060/100000 progress=1.1% loss=0.732422 lr=2.118000e-06 elapsed=00:24:00 avg_step=1.36s recent_step=1.26s eta=37:20:27 +2026-08-25 04:19:41 | INFO | train rank=0 | step=1070/100000 progress=1.1% loss=0.700684 lr=2.138000e-06 elapsed=00:24:11 avg_step=1.36s recent_step=1.15s eta=37:16:58 +2026-08-25 04:19:53 | INFO | train rank=0 | step=1080/100000 progress=1.1% loss=0.739746 lr=2.158000e-06 elapsed=00:24:23 avg_step=1.35s recent_step=1.16s eta=37:13:40 +2026-08-25 04:20:06 | INFO | train rank=0 | step=1090/100000 progress=1.1% loss=0.772949 lr=2.178000e-06 elapsed=00:24:35 avg_step=1.35s recent_step=1.27s eta=37:12:13 +2026-08-25 04:20:22 | INFO | train rank=0 | step=1100/100000 progress=1.1% loss=0.684082 lr=2.198000e-06 elapsed=00:24:51 avg_step=1.36s recent_step=1.60s eta=37:15:39 +2026-08-25 04:20:33 | INFO | train rank=0 | step=1110/100000 progress=1.1% loss=0.665527 lr=2.218000e-06 elapsed=00:25:03 avg_step=1.35s recent_step=1.19s eta=37:12:54 +2026-08-25 04:20:46 | INFO | train rank=0 | step=1120/100000 progress=1.1% loss=0.668457 lr=2.238000e-06 elapsed=00:25:15 avg_step=1.35s recent_step=1.21s eta=37:10:33 +2026-08-25 04:21:00 | INFO | train rank=0 | step=1130/100000 progress=1.1% loss=0.689941 lr=2.258000e-06 elapsed=00:25:30 avg_step=1.35s recent_step=1.48s eta=37:12:09 +2026-08-25 04:21:15 | INFO | train rank=0 | step=1140/100000 progress=1.1% loss=0.690430 lr=2.278000e-06 elapsed=00:25:45 avg_step=1.36s recent_step=1.51s eta=37:14:14 +2026-08-25 04:21:30 | INFO | train rank=0 | step=1150/100000 progress=1.1% loss=0.654297 lr=2.298000e-06 elapsed=00:26:00 avg_step=1.36s recent_step=1.43s eta=37:15:08 +2026-08-25 04:21:42 | INFO | train rank=0 | step=1160/100000 progress=1.2% loss=0.644531 lr=2.318000e-06 elapsed=00:26:12 avg_step=1.36s recent_step=1.20s eta=37:12:42 +2026-08-25 04:21:54 | INFO | train rank=0 | step=1170/100000 progress=1.2% loss=0.725586 lr=2.338000e-06 elapsed=00:26:24 avg_step=1.35s recent_step=1.18s eta=37:10:03 +2026-08-25 04:22:06 | INFO | train rank=0 | step=1180/100000 progress=1.2% loss=0.578613 lr=2.358000e-06 elapsed=00:26:36 avg_step=1.35s recent_step=1.20s eta=37:07:42 +2026-08-25 04:22:17 | INFO | train rank=0 | step=1190/100000 progress=1.2% loss=0.568359 lr=2.378000e-06 elapsed=00:26:47 avg_step=1.35s recent_step=1.18s eta=37:05:01 +2026-08-25 04:22:30 | INFO | train rank=0 | step=1200/100000 progress=1.2% loss=0.565918 lr=2.398000e-06 elapsed=00:27:00 avg_step=1.35s recent_step=1.23s eta=37:03:06 +2026-08-25 04:22:43 | INFO | train rank=0 | step=1210/100000 progress=1.2% loss=0.582031 lr=2.418000e-06 elapsed=00:27:13 avg_step=1.35s recent_step=1.35s eta=37:02:54 +2026-08-25 04:22:55 | INFO | train rank=0 | step=1220/100000 progress=1.2% loss=0.623047 lr=2.438000e-06 elapsed=00:27:25 avg_step=1.35s recent_step=1.20s eta=37:00:43 +2026-08-25 04:23:08 | INFO | train rank=0 | step=1230/100000 progress=1.2% loss=0.588379 lr=2.458000e-06 elapsed=00:27:37 avg_step=1.35s recent_step=1.22s eta=36:58:50 +2026-08-25 04:23:19 | INFO | train rank=0 | step=1240/100000 progress=1.2% loss=0.526611 lr=2.478000e-06 elapsed=00:27:49 avg_step=1.35s recent_step=1.19s eta=36:56:34 +2026-08-25 04:23:36 | INFO | train rank=0 | step=1250/100000 progress=1.2% loss=0.500488 lr=2.498000e-06 elapsed=00:28:06 avg_step=1.35s recent_step=1.65s eta=37:00:22 +2026-08-25 04:23:51 | INFO | train rank=0 | step=1260/100000 progress=1.3% loss=0.479492 lr=2.518000e-06 elapsed=00:28:20 avg_step=1.35s recent_step=1.46s eta=37:01:37 +2026-08-25 04:24:03 | INFO | train rank=0 | step=1270/100000 progress=1.3% loss=0.544434 lr=2.538000e-06 elapsed=00:28:33 avg_step=1.35s recent_step=1.22s eta=36:59:43 +2026-08-25 04:24:15 | INFO | train rank=0 | step=1280/100000 progress=1.3% loss=0.531250 lr=2.558000e-06 elapsed=00:28:45 avg_step=1.35s recent_step=1.22s eta=36:57:51 +2026-08-25 04:24:28 | INFO | train rank=0 | step=1290/100000 progress=1.3% loss=0.527832 lr=2.578000e-06 elapsed=00:28:58 avg_step=1.35s recent_step=1.35s eta=36:57:36 +2026-08-25 04:24:41 | INFO | train rank=0 | step=1300/100000 progress=1.3% loss=0.573730 lr=2.598000e-06 elapsed=00:29:11 avg_step=1.35s recent_step=1.28s eta=36:56:31 +2026-08-25 04:24:55 | INFO | train rank=0 | step=1310/100000 progress=1.3% loss=0.441650 lr=2.618000e-06 elapsed=00:29:25 avg_step=1.35s recent_step=1.39s eta=36:56:50 +2026-08-25 04:25:09 | INFO | train rank=0 | step=1320/100000 progress=1.3% loss=0.460693 lr=2.638000e-06 elapsed=00:29:39 avg_step=1.35s recent_step=1.42s eta=36:57:31 +2026-08-25 04:25:22 | INFO | train rank=0 | step=1330/100000 progress=1.3% loss=0.449463 lr=2.658000e-06 elapsed=00:29:52 avg_step=1.35s recent_step=1.25s eta=36:56:07 +2026-08-25 04:25:35 | INFO | train rank=0 | step=1340/100000 progress=1.3% loss=0.401123 lr=2.678000e-06 elapsed=00:30:05 avg_step=1.35s recent_step=1.35s eta=36:55:55 +2026-08-25 04:25:47 | INFO | train rank=0 | step=1350/100000 progress=1.4% loss=0.403809 lr=2.698000e-06 elapsed=00:30:17 avg_step=1.35s recent_step=1.19s eta=36:53:44 +2026-08-25 04:25:59 | INFO | train rank=0 | step=1360/100000 progress=1.4% loss=0.451660 lr=2.718000e-06 elapsed=00:30:29 avg_step=1.35s recent_step=1.18s eta=36:51:27 +2026-08-25 04:26:11 | INFO | train rank=0 | step=1370/100000 progress=1.4% loss=0.454102 lr=2.738000e-06 elapsed=00:30:41 avg_step=1.34s recent_step=1.24s eta=36:49:56 +2026-08-25 04:26:24 | INFO | train rank=0 | step=1380/100000 progress=1.4% loss=0.425049 lr=2.758000e-06 elapsed=00:30:54 avg_step=1.34s recent_step=1.24s eta=36:48:30 +2026-08-25 04:26:35 | INFO | train rank=0 | step=1390/100000 progress=1.4% loss=0.532471 lr=2.778000e-06 elapsed=00:31:05 avg_step=1.34s recent_step=1.16s eta=36:46:10 +2026-08-25 04:26:50 | INFO | train rank=0 | step=1400/100000 progress=1.4% loss=0.464355 lr=2.798000e-06 elapsed=00:31:20 avg_step=1.34s recent_step=1.42s eta=36:46:52 +2026-08-25 04:27:02 | INFO | train rank=0 | step=1410/100000 progress=1.4% loss=0.398438 lr=2.818000e-06 elapsed=00:31:32 avg_step=1.34s recent_step=1.23s eta=36:45:17 +2026-08-25 04:27:15 | INFO | train rank=0 | step=1420/100000 progress=1.4% loss=0.368896 lr=2.838000e-06 elapsed=00:31:45 avg_step=1.34s recent_step=1.28s eta=36:44:21 +2026-08-25 04:27:28 | INFO | train rank=0 | step=1430/100000 progress=1.4% loss=0.352783 lr=2.858000e-06 elapsed=00:31:58 avg_step=1.34s recent_step=1.33s eta=36:44:01 +2026-08-25 04:27:41 | INFO | train rank=0 | step=1440/100000 progress=1.4% loss=0.332031 lr=2.878000e-06 elapsed=00:32:11 avg_step=1.34s recent_step=1.32s eta=36:43:35 +2026-08-25 04:27:54 | INFO | train rank=0 | step=1450/100000 progress=1.5% loss=0.323730 lr=2.898000e-06 elapsed=00:32:23 avg_step=1.34s recent_step=1.22s eta=36:41:58 +2026-08-25 04:28:08 | INFO | train rank=0 | step=1460/100000 progress=1.5% loss=0.453369 lr=2.918000e-06 elapsed=00:32:38 avg_step=1.34s recent_step=1.45s eta=36:42:55 +2026-08-25 04:28:21 | INFO | train rank=0 | step=1470/100000 progress=1.5% loss=0.330566 lr=2.938000e-06 elapsed=00:32:51 avg_step=1.34s recent_step=1.33s eta=36:42:31 +2026-08-25 04:28:34 | INFO | train rank=0 | step=1480/100000 progress=1.5% loss=0.315430 lr=2.958000e-06 elapsed=00:33:04 avg_step=1.34s recent_step=1.29s eta=36:41:44 +2026-08-25 04:28:51 | INFO | train rank=0 | step=1490/100000 progress=1.5% loss=0.394775 lr=2.978000e-06 elapsed=00:33:21 avg_step=1.34s recent_step=1.70s eta=36:45:29 +2026-08-25 04:29:03 | INFO | train rank=0 | step=1500/100000 progress=1.5% loss=0.336670 lr=2.998000e-06 elapsed=00:33:33 avg_step=1.34s recent_step=1.16s eta=36:43:13 +2026-08-25 04:29:15 | INFO | train rank=0 | step=1510/100000 progress=1.5% loss=0.286865 lr=3.018000e-06 elapsed=00:33:45 avg_step=1.34s recent_step=1.21s eta=36:41:32 +2026-08-25 04:29:26 | INFO | train rank=0 | step=1520/100000 progress=1.5% loss=0.348877 lr=3.038000e-06 elapsed=00:33:56 avg_step=1.34s recent_step=1.13s eta=36:39:01 +2026-08-25 04:29:45 | INFO | train rank=0 | step=1530/100000 progress=1.5% loss=0.335205 lr=3.058000e-06 elapsed=00:34:15 avg_step=1.34s recent_step=1.90s eta=36:44:47 +2026-08-25 04:29:57 | INFO | train rank=0 | step=1540/100000 progress=1.5% loss=0.270142 lr=3.078000e-06 elapsed=00:34:27 avg_step=1.34s recent_step=1.24s eta=36:43:26 +2026-08-25 04:30:13 | INFO | train rank=0 | step=1550/100000 progress=1.6% loss=0.296875 lr=3.098000e-06 elapsed=00:34:43 avg_step=1.34s recent_step=1.57s eta=36:45:36 +2026-08-25 04:30:25 | INFO | train rank=0 | step=1560/100000 progress=1.6% loss=0.242188 lr=3.118000e-06 elapsed=00:34:55 avg_step=1.34s recent_step=1.17s eta=36:43:33 +2026-08-25 04:30:37 | INFO | train rank=0 | step=1570/100000 progress=1.6% loss=0.303467 lr=3.138000e-06 elapsed=00:35:06 avg_step=1.34s recent_step=1.18s eta=36:41:35 +2026-08-25 04:30:48 | INFO | train rank=0 | step=1580/100000 progress=1.6% loss=0.330811 lr=3.158000e-06 elapsed=00:35:18 avg_step=1.34s recent_step=1.13s eta=36:39:12 +2026-08-25 04:31:01 | INFO | train rank=0 | step=1590/100000 progress=1.6% loss=0.347168 lr=3.178000e-06 elapsed=00:35:31 avg_step=1.34s recent_step=1.30s eta=36:38:34 +2026-08-25 04:31:15 | INFO | train rank=0 | step=1600/100000 progress=1.6% loss=0.287842 lr=3.198000e-06 elapsed=00:35:45 avg_step=1.34s recent_step=1.44s eta=36:39:21 +2026-08-25 04:31:27 | INFO | train rank=0 | step=1610/100000 progress=1.6% loss=0.271362 lr=3.218000e-06 elapsed=00:35:57 avg_step=1.34s recent_step=1.13s eta=36:37:00 +2026-08-25 04:31:41 | INFO | train rank=0 | step=1620/100000 progress=1.6% loss=0.246948 lr=3.238000e-06 elapsed=00:36:11 avg_step=1.34s recent_step=1.44s eta=36:37:50 +2026-08-25 04:31:53 | INFO | train rank=0 | step=1630/100000 progress=1.6% loss=0.226074 lr=3.258000e-06 elapsed=00:36:23 avg_step=1.34s recent_step=1.21s eta=36:36:16 +2026-08-25 04:32:07 | INFO | train rank=0 | step=1640/100000 progress=1.6% loss=0.297607 lr=3.278000e-06 elapsed=00:36:37 avg_step=1.34s recent_step=1.40s eta=36:36:38 +2026-08-25 04:32:24 | INFO | train rank=0 | step=1650/100000 progress=1.7% loss=0.238403 lr=3.298000e-06 elapsed=00:36:54 avg_step=1.34s recent_step=1.73s eta=36:40:15 +2026-08-25 04:32:36 | INFO | train rank=0 | step=1660/100000 progress=1.7% loss=0.275391 lr=3.318000e-06 elapsed=00:37:06 avg_step=1.34s recent_step=1.14s eta=36:38:00 +2026-08-25 04:32:48 | INFO | train rank=0 | step=1670/100000 progress=1.7% loss=0.244751 lr=3.338000e-06 elapsed=00:37:18 avg_step=1.34s recent_step=1.26s eta=36:36:59 +2026-08-25 04:33:00 | INFO | train rank=0 | step=1680/100000 progress=1.7% loss=0.295776 lr=3.358000e-06 elapsed=00:37:30 avg_step=1.34s recent_step=1.18s eta=36:35:12 +2026-08-25 04:33:12 | INFO | train rank=0 | step=1690/100000 progress=1.7% loss=0.256836 lr=3.378000e-06 elapsed=00:37:42 avg_step=1.34s recent_step=1.21s eta=36:33:42 +2026-08-25 04:33:25 | INFO | train rank=0 | step=1700/100000 progress=1.7% loss=0.220947 lr=3.398000e-06 elapsed=00:37:55 avg_step=1.34s recent_step=1.24s eta=36:32:32 +2026-08-25 04:33:37 | INFO | train rank=0 | step=1710/100000 progress=1.7% loss=0.343018 lr=3.418000e-06 elapsed=00:38:07 avg_step=1.34s recent_step=1.22s eta=36:31:09 +2026-08-25 04:33:49 | INFO | train rank=0 | step=1720/100000 progress=1.7% loss=0.236694 lr=3.438000e-06 elapsed=00:38:19 avg_step=1.34s recent_step=1.23s eta=36:29:57 +2026-08-25 04:34:06 | INFO | train rank=0 | step=1730/100000 progress=1.7% loss=0.233887 lr=3.458000e-06 elapsed=00:38:36 avg_step=1.34s recent_step=1.66s eta=36:32:45 +2026-08-25 04:34:18 | INFO | train rank=0 | step=1740/100000 progress=1.7% loss=0.291504 lr=3.478000e-06 elapsed=00:38:48 avg_step=1.34s recent_step=1.21s eta=36:31:17 +2026-08-25 04:34:31 | INFO | train rank=0 | step=1750/100000 progress=1.8% loss=0.225586 lr=3.498000e-06 elapsed=00:39:01 avg_step=1.34s recent_step=1.28s eta=36:30:33 +2026-08-25 04:34:42 | INFO | train rank=0 | step=1760/100000 progress=1.8% loss=0.252319 lr=3.518000e-06 elapsed=00:39:12 avg_step=1.34s recent_step=1.17s eta=36:28:46 +2026-08-25 04:34:56 | INFO | train rank=0 | step=1770/100000 progress=1.8% loss=0.247314 lr=3.538000e-06 elapsed=00:39:26 avg_step=1.34s recent_step=1.40s eta=36:29:06 +2026-08-25 04:35:09 | INFO | train rank=0 | step=1780/100000 progress=1.8% loss=0.234253 lr=3.558000e-06 elapsed=00:39:39 avg_step=1.34s recent_step=1.24s eta=36:28:00 +2026-08-25 04:35:20 | INFO | train rank=0 | step=1790/100000 progress=1.8% loss=0.203125 lr=3.578000e-06 elapsed=00:39:50 avg_step=1.34s recent_step=1.16s eta=36:26:07 +2026-08-25 04:35:34 | INFO | train rank=0 | step=1800/100000 progress=1.8% loss=0.273438 lr=3.598000e-06 elapsed=00:40:04 avg_step=1.34s recent_step=1.33s eta=36:25:53 +2026-08-25 04:35:45 | INFO | train rank=0 | step=1810/100000 progress=1.8% loss=0.259155 lr=3.618000e-06 elapsed=00:40:15 avg_step=1.33s recent_step=1.18s eta=36:24:14 +2026-08-25 04:35:57 | INFO | train rank=0 | step=1820/100000 progress=1.8% loss=0.287476 lr=3.638000e-06 elapsed=00:40:27 avg_step=1.33s recent_step=1.21s eta=36:22:52 +2026-08-25 04:36:10 | INFO | train rank=0 | step=1830/100000 progress=1.8% loss=0.235718 lr=3.658000e-06 elapsed=00:40:40 avg_step=1.33s recent_step=1.25s eta=36:21:56 +2026-08-25 04:36:22 | INFO | train rank=0 | step=1840/100000 progress=1.8% loss=0.243530 lr=3.678000e-06 elapsed=00:40:52 avg_step=1.33s recent_step=1.17s eta=36:20:17 +2026-08-25 04:36:34 | INFO | train rank=0 | step=1850/100000 progress=1.8% loss=0.279297 lr=3.698000e-06 elapsed=00:41:04 avg_step=1.33s recent_step=1.22s eta=36:19:05 +2026-08-25 04:36:46 | INFO | train rank=0 | step=1860/100000 progress=1.9% loss=0.174194 lr=3.718000e-06 elapsed=00:41:16 avg_step=1.33s recent_step=1.24s eta=36:18:03 +2026-08-25 04:36:59 | INFO | train rank=0 | step=1870/100000 progress=1.9% loss=0.195190 lr=3.738000e-06 elapsed=00:41:28 avg_step=1.33s recent_step=1.22s eta=36:16:51 +2026-08-25 04:37:13 | INFO | train rank=0 | step=1880/100000 progress=1.9% loss=0.312622 lr=3.758000e-06 elapsed=00:41:43 avg_step=1.33s recent_step=1.43s eta=36:17:31 +2026-08-25 04:37:29 | INFO | train rank=0 | step=1890/100000 progress=1.9% loss=0.191772 lr=3.778000e-06 elapsed=00:41:59 avg_step=1.33s recent_step=1.58s eta=36:19:25 +2026-08-25 04:37:41 | INFO | train rank=0 | step=1900/100000 progress=1.9% loss=0.192749 lr=3.798000e-06 elapsed=00:42:10 avg_step=1.33s recent_step=1.19s eta=36:17:58 +2026-08-25 04:37:55 | INFO | train rank=0 | step=1910/100000 progress=1.9% loss=0.232300 lr=3.818000e-06 elapsed=00:42:25 avg_step=1.33s recent_step=1.41s eta=36:18:23 +2026-08-25 04:38:10 | INFO | train rank=0 | step=1920/100000 progress=1.9% loss=0.219971 lr=3.838000e-06 elapsed=00:42:40 avg_step=1.33s recent_step=1.55s eta=36:20:03 +2026-08-25 04:38:23 | INFO | train rank=0 | step=1930/100000 progress=1.9% loss=0.141541 lr=3.858000e-06 elapsed=00:42:53 avg_step=1.33s recent_step=1.33s eta=36:19:48 +2026-08-25 04:38:36 | INFO | train rank=0 | step=1940/100000 progress=1.9% loss=0.171875 lr=3.878000e-06 elapsed=00:43:06 avg_step=1.33s recent_step=1.21s eta=36:18:33 +2026-08-25 04:38:49 | INFO | train rank=0 | step=1950/100000 progress=1.9% loss=0.168823 lr=3.898000e-06 elapsed=00:43:19 avg_step=1.33s recent_step=1.35s eta=36:18:28 +2026-08-25 04:39:05 | INFO | train rank=0 | step=1960/100000 progress=2.0% loss=0.222412 lr=3.918000e-06 elapsed=00:43:35 avg_step=1.33s recent_step=1.58s eta=36:20:18 +2026-08-25 04:39:18 | INFO | train rank=0 | step=1970/100000 progress=2.0% loss=0.150696 lr=3.938000e-06 elapsed=00:43:48 avg_step=1.33s recent_step=1.33s eta=36:20:04 +2026-08-25 04:39:33 | INFO | train rank=0 | step=1980/100000 progress=2.0% loss=0.151917 lr=3.958000e-06 elapsed=00:44:03 avg_step=1.33s recent_step=1.45s eta=36:20:47 +2026-08-25 04:39:47 | INFO | train rank=0 | step=1990/100000 progress=2.0% loss=0.237915 lr=3.978000e-06 elapsed=00:44:17 avg_step=1.34s recent_step=1.42s eta=36:21:13 +2026-08-25 04:40:00 | INFO | train rank=0 | step=2000/100000 progress=2.0% loss=0.245850 lr=3.998000e-06 elapsed=00:44:30 avg_step=1.34s recent_step=1.32s eta=36:20:52 +2026-08-25 04:40:12 | INFO | train rank=0 | step=2010/100000 progress=2.0% loss=0.177856 lr=4.018000e-06 elapsed=00:44:42 avg_step=1.33s recent_step=1.21s eta=36:19:38 +2026-08-25 04:40:25 | INFO | train rank=0 | step=2020/100000 progress=2.0% loss=0.150330 lr=4.038000e-06 elapsed=00:44:55 avg_step=1.33s recent_step=1.28s eta=36:19:00 +2026-08-25 04:40:37 | INFO | train rank=0 | step=2030/100000 progress=2.0% loss=0.203125 lr=4.058000e-06 elapsed=00:45:07 avg_step=1.33s recent_step=1.18s eta=36:17:34 +2026-08-25 04:40:51 | INFO | train rank=0 | step=2040/100000 progress=2.0% loss=0.164551 lr=4.078000e-06 elapsed=00:45:21 avg_step=1.33s recent_step=1.43s eta=36:18:08 +2026-08-25 04:41:04 | INFO | train rank=0 | step=2050/100000 progress=2.1% loss=0.160645 lr=4.098000e-06 elapsed=00:45:34 avg_step=1.33s recent_step=1.26s eta=36:17:21 +2026-08-25 04:41:19 | INFO | train rank=0 | step=2060/100000 progress=2.1% loss=0.222900 lr=4.118000e-06 elapsed=00:45:49 avg_step=1.33s recent_step=1.55s eta=36:18:49 +2026-08-25 04:41:34 | INFO | train rank=0 | step=2070/100000 progress=2.1% loss=0.172852 lr=4.138000e-06 elapsed=00:46:04 avg_step=1.34s recent_step=1.44s eta=36:19:27 +2026-08-25 04:41:46 | INFO | train rank=0 | step=2080/100000 progress=2.1% loss=0.204346 lr=4.158000e-06 elapsed=00:46:16 avg_step=1.33s recent_step=1.20s eta=36:18:10 +2026-08-25 04:41:58 | INFO | train rank=0 | step=2090/100000 progress=2.1% loss=0.185730 lr=4.178000e-06 elapsed=00:46:28 avg_step=1.33s recent_step=1.20s eta=36:16:55 +2026-08-25 04:42:10 | INFO | train rank=0 | step=2100/100000 progress=2.1% loss=0.172974 lr=4.198000e-06 elapsed=00:46:39 avg_step=1.33s recent_step=1.18s eta=36:15:30 +2026-08-25 04:42:22 | INFO | train rank=0 | step=2110/100000 progress=2.1% loss=0.170288 lr=4.218000e-06 elapsed=00:46:52 avg_step=1.33s recent_step=1.25s eta=36:14:39 +2026-08-25 04:42:35 | INFO | train rank=0 | step=2120/100000 progress=2.1% loss=0.182007 lr=4.238000e-06 elapsed=00:47:04 avg_step=1.33s recent_step=1.25s eta=36:13:45 +2026-08-25 04:42:48 | INFO | train rank=0 | step=2130/100000 progress=2.1% loss=0.114258 lr=4.258000e-06 elapsed=00:47:18 avg_step=1.33s recent_step=1.31s eta=36:13:22 +2026-08-25 04:43:00 | INFO | train rank=0 | step=2140/100000 progress=2.1% loss=0.144226 lr=4.278000e-06 elapsed=00:47:30 avg_step=1.33s recent_step=1.25s eta=36:12:32 +2026-08-25 04:43:15 | INFO | train rank=0 | step=2150/100000 progress=2.1% loss=0.185303 lr=4.298000e-06 elapsed=00:47:45 avg_step=1.33s recent_step=1.45s eta=36:13:15 +2026-08-25 04:43:27 | INFO | train rank=0 | step=2160/100000 progress=2.2% loss=0.281982 lr=4.318000e-06 elapsed=00:47:57 avg_step=1.33s recent_step=1.22s eta=36:12:09 +2026-08-25 04:43:39 | INFO | train rank=0 | step=2170/100000 progress=2.2% loss=0.152527 lr=4.338000e-06 elapsed=00:48:09 avg_step=1.33s recent_step=1.20s eta=36:10:58 +2026-08-25 04:43:51 | INFO | train rank=0 | step=2180/100000 progress=2.2% loss=0.111206 lr=4.358000e-06 elapsed=00:48:21 avg_step=1.33s recent_step=1.17s eta=36:09:34 +2026-08-25 04:44:03 | INFO | train rank=0 | step=2190/100000 progress=2.2% loss=0.158203 lr=4.378000e-06 elapsed=00:48:33 avg_step=1.33s recent_step=1.22s eta=36:08:33 +2026-08-25 04:44:15 | INFO | train rank=0 | step=2200/100000 progress=2.2% loss=0.115540 lr=4.398000e-06 elapsed=00:48:45 avg_step=1.33s recent_step=1.23s eta=36:07:35 +2026-08-25 04:44:27 | INFO | train rank=0 | step=2210/100000 progress=2.2% loss=0.145508 lr=4.418000e-06 elapsed=00:48:57 avg_step=1.33s recent_step=1.17s eta=36:06:11 +2026-08-25 04:44:39 | INFO | train rank=0 | step=2220/100000 progress=2.2% loss=0.123718 lr=4.438000e-06 elapsed=00:49:09 avg_step=1.33s recent_step=1.23s eta=36:05:15 +2026-08-25 04:44:51 | INFO | train rank=0 | step=2230/100000 progress=2.2% loss=0.187378 lr=4.458000e-06 elapsed=00:49:21 avg_step=1.33s recent_step=1.22s eta=36:04:16 +2026-08-25 04:45:08 | INFO | train rank=0 | step=2240/100000 progress=2.2% loss=0.183472 lr=4.478000e-06 elapsed=00:49:38 avg_step=1.33s recent_step=1.65s eta=36:06:23 +2026-08-25 04:45:20 | INFO | train rank=0 | step=2250/100000 progress=2.2% loss=0.152039 lr=4.498000e-06 elapsed=00:49:50 avg_step=1.33s recent_step=1.22s eta=36:05:23 +2026-08-25 04:45:34 | INFO | train rank=0 | step=2260/100000 progress=2.3% loss=0.206177 lr=4.518000e-06 elapsed=00:50:04 avg_step=1.33s recent_step=1.43s eta=36:05:52 +2026-08-25 04:45:47 | INFO | train rank=0 | step=2270/100000 progress=2.3% loss=0.202271 lr=4.538000e-06 elapsed=00:50:17 avg_step=1.33s recent_step=1.22s eta=36:04:52 +2026-08-25 04:45:59 | INFO | train rank=0 | step=2280/100000 progress=2.3% loss=0.119507 lr=4.558000e-06 elapsed=00:50:29 avg_step=1.33s recent_step=1.27s eta=36:04:15 +2026-08-25 04:46:12 | INFO | train rank=0 | step=2290/100000 progress=2.3% loss=0.182861 lr=4.578000e-06 elapsed=00:50:41 avg_step=1.33s recent_step=1.21s eta=36:03:13 +2026-08-25 04:46:28 | INFO | train rank=0 | step=2300/100000 progress=2.3% loss=0.213745 lr=4.598000e-06 elapsed=00:50:58 avg_step=1.33s recent_step=1.61s eta=36:04:58 +2026-08-25 04:46:39 | INFO | train rank=0 | step=2310/100000 progress=2.3% loss=0.204956 lr=4.618000e-06 elapsed=00:51:09 avg_step=1.33s recent_step=1.19s eta=36:03:44 +2026-08-25 04:46:56 | INFO | train rank=0 | step=2320/100000 progress=2.3% loss=0.190674 lr=4.638000e-06 elapsed=00:51:25 avg_step=1.33s recent_step=1.61s eta=36:05:28 +2026-08-25 04:47:09 | INFO | train rank=0 | step=2330/100000 progress=2.3% loss=0.120667 lr=4.658000e-06 elapsed=00:51:39 avg_step=1.33s recent_step=1.37s eta=36:05:33 +2026-08-25 04:47:21 | INFO | train rank=0 | step=2340/100000 progress=2.3% loss=0.135498 lr=4.678000e-06 elapsed=00:51:51 avg_step=1.33s recent_step=1.18s eta=36:04:15 +2026-08-25 04:47:33 | INFO | train rank=0 | step=2350/100000 progress=2.4% loss=0.239380 lr=4.698000e-06 elapsed=00:52:03 avg_step=1.33s recent_step=1.24s eta=36:03:26 +2026-08-25 04:47:46 | INFO | train rank=0 | step=2360/100000 progress=2.4% loss=0.140259 lr=4.718000e-06 elapsed=00:52:16 avg_step=1.33s recent_step=1.27s eta=36:02:46 +2026-08-25 04:48:01 | INFO | train rank=0 | step=2370/100000 progress=2.4% loss=0.209229 lr=4.738000e-06 elapsed=00:52:30 avg_step=1.33s recent_step=1.44s eta=36:03:20 +2026-08-25 04:48:13 | INFO | train rank=0 | step=2380/100000 progress=2.4% loss=0.146301 lr=4.758000e-06 elapsed=00:52:43 avg_step=1.33s recent_step=1.24s eta=36:02:28 +2026-08-25 04:48:25 | INFO | train rank=0 | step=2390/100000 progress=2.4% loss=0.171021 lr=4.778000e-06 elapsed=00:52:55 avg_step=1.33s recent_step=1.19s eta=36:01:19 +2026-08-25 04:48:43 | INFO | train rank=0 | step=2400/100000 progress=2.4% loss=0.139893 lr=4.798000e-06 elapsed=00:53:13 avg_step=1.33s recent_step=1.79s eta=36:04:12 +2026-08-25 04:48:55 | INFO | train rank=0 | step=2410/100000 progress=2.4% loss=0.106750 lr=4.818000e-06 elapsed=00:53:24 avg_step=1.33s recent_step=1.18s eta=36:02:59 +2026-08-25 04:49:09 | INFO | train rank=0 | step=2420/100000 progress=2.4% loss=0.250366 lr=4.838000e-06 elapsed=00:53:39 avg_step=1.33s recent_step=1.45s eta=36:03:35 +2026-08-25 04:49:21 | INFO | train rank=0 | step=2430/100000 progress=2.4% loss=0.107605 lr=4.858000e-06 elapsed=00:53:51 avg_step=1.33s recent_step=1.21s eta=36:02:34 +2026-08-25 04:49:33 | INFO | train rank=0 | step=2440/100000 progress=2.4% loss=0.121948 lr=4.878000e-06 elapsed=00:54:03 avg_step=1.33s recent_step=1.18s eta=36:01:19 +2026-08-25 04:49:45 | INFO | train rank=0 | step=2450/100000 progress=2.5% loss=0.122803 lr=4.898000e-06 elapsed=00:54:15 avg_step=1.33s recent_step=1.23s eta=36:00:26 +2026-08-25 04:50:00 | INFO | train rank=0 | step=2460/100000 progress=2.5% loss=0.152039 lr=4.918000e-06 elapsed=00:54:29 avg_step=1.33s recent_step=1.43s eta=36:00:53 +2026-08-25 04:50:11 | INFO | train rank=0 | step=2470/100000 progress=2.5% loss=0.127197 lr=4.938000e-06 elapsed=00:54:41 avg_step=1.33s recent_step=1.20s eta=35:59:47 +2026-08-25 04:50:25 | INFO | train rank=0 | step=2480/100000 progress=2.5% loss=0.129333 lr=4.958000e-06 elapsed=00:54:55 avg_step=1.33s recent_step=1.32s eta=35:59:31 +2026-08-25 04:50:39 | INFO | train rank=0 | step=2490/100000 progress=2.5% loss=0.119507 lr=4.978000e-06 elapsed=00:55:09 avg_step=1.33s recent_step=1.39s eta=35:59:44 +2026-08-25 04:50:51 | INFO | train rank=0 | step=2500/100000 progress=2.5% loss=0.127197 lr=4.998000e-06 elapsed=00:55:21 avg_step=1.33s recent_step=1.23s eta=35:58:52 +2026-08-25 04:51:04 | INFO | train rank=0 | step=2510/100000 progress=2.5% loss=0.126221 lr=5.018000e-06 elapsed=00:55:34 avg_step=1.33s recent_step=1.33s eta=35:58:39 +2026-08-25 04:51:18 | INFO | train rank=0 | step=2520/100000 progress=2.5% loss=0.105957 lr=5.038000e-06 elapsed=00:55:48 avg_step=1.33s recent_step=1.42s eta=35:59:00 +2026-08-25 04:51:30 | INFO | train rank=0 | step=2530/100000 progress=2.5% loss=0.221069 lr=5.058000e-06 elapsed=00:56:00 avg_step=1.33s recent_step=1.19s eta=35:57:53 +2026-08-25 04:51:45 | INFO | train rank=0 | step=2540/100000 progress=2.5% loss=0.183350 lr=5.078000e-06 elapsed=00:56:14 avg_step=1.33s recent_step=1.42s eta=35:58:14 +2026-08-25 04:51:57 | INFO | train rank=0 | step=2550/100000 progress=2.5% loss=0.123901 lr=5.098000e-06 elapsed=00:56:27 avg_step=1.33s recent_step=1.22s eta=35:57:19 +2026-08-25 04:52:10 | INFO | train rank=0 | step=2560/100000 progress=2.6% loss=0.131897 lr=5.118000e-06 elapsed=00:56:40 avg_step=1.33s recent_step=1.37s eta=35:57:21 +2026-08-25 04:52:25 | INFO | train rank=0 | step=2570/100000 progress=2.6% loss=0.135193 lr=5.138000e-06 elapsed=00:56:55 avg_step=1.33s recent_step=1.47s eta=35:58:02 +2026-08-25 04:52:39 | INFO | train rank=0 | step=2580/100000 progress=2.6% loss=0.126892 lr=5.158000e-06 elapsed=00:57:09 avg_step=1.33s recent_step=1.41s eta=35:58:18 +2026-08-25 04:52:52 | INFO | train rank=0 | step=2590/100000 progress=2.6% loss=0.129456 lr=5.178000e-06 elapsed=00:57:22 avg_step=1.33s recent_step=1.25s eta=35:57:36 +2026-08-25 04:53:08 | INFO | train rank=0 | step=2600/100000 progress=2.6% loss=0.135376 lr=5.198000e-06 elapsed=00:57:38 avg_step=1.33s recent_step=1.63s eta=35:59:15 +2026-08-25 04:53:20 | INFO | train rank=0 | step=2610/100000 progress=2.6% loss=0.136841 lr=5.218000e-06 elapsed=00:57:50 avg_step=1.33s recent_step=1.23s eta=35:58:22 +2026-08-25 04:53:36 | INFO | train rank=0 | step=2620/100000 progress=2.6% loss=0.118042 lr=5.238000e-06 elapsed=00:58:06 avg_step=1.33s recent_step=1.58s eta=35:59:42 +2026-08-25 04:53:48 | INFO | train rank=0 | step=2630/100000 progress=2.6% loss=0.198364 lr=5.258000e-06 elapsed=00:58:18 avg_step=1.33s recent_step=1.21s eta=35:58:46 +2026-08-25 04:54:00 | INFO | train rank=0 | step=2640/100000 progress=2.6% loss=0.118591 lr=5.278000e-06 elapsed=00:58:30 avg_step=1.33s recent_step=1.22s eta=35:57:53 +2026-08-25 04:54:12 | INFO | train rank=0 | step=2650/100000 progress=2.6% loss=0.151672 lr=5.298000e-06 elapsed=00:58:42 avg_step=1.33s recent_step=1.20s eta=35:56:52 +2026-08-25 04:54:24 | INFO | train rank=0 | step=2660/100000 progress=2.7% loss=0.189453 lr=5.318000e-06 elapsed=00:58:54 avg_step=1.33s recent_step=1.17s eta=35:55:42 +2026-08-25 04:54:36 | INFO | train rank=0 | step=2670/100000 progress=2.7% loss=0.113892 lr=5.338000e-06 elapsed=00:59:06 avg_step=1.33s recent_step=1.22s eta=35:54:50 +2026-08-25 04:54:48 | INFO | train rank=0 | step=2680/100000 progress=2.7% loss=0.150330 lr=5.358000e-06 elapsed=00:59:18 avg_step=1.33s recent_step=1.18s eta=35:53:44 +2026-08-25 04:55:00 | INFO | train rank=0 | step=2690/100000 progress=2.7% loss=0.120300 lr=5.378000e-06 elapsed=00:59:30 avg_step=1.33s recent_step=1.20s eta=35:52:45 +2026-08-25 04:55:12 | INFO | train rank=0 | step=2700/100000 progress=2.7% loss=0.124329 lr=5.398000e-06 elapsed=00:59:42 avg_step=1.33s recent_step=1.20s eta=35:51:46 +2026-08-25 04:55:25 | INFO | train rank=0 | step=2710/100000 progress=2.7% loss=0.152588 lr=5.418000e-06 elapsed=00:59:55 avg_step=1.33s recent_step=1.31s eta=35:51:28 +2026-08-25 04:55:37 | INFO | train rank=0 | step=2720/100000 progress=2.7% loss=0.101196 lr=5.438000e-06 elapsed=01:00:07 avg_step=1.33s recent_step=1.18s eta=35:50:23 +2026-08-25 04:55:53 | INFO | train rank=0 | step=2730/100000 progress=2.7% loss=0.138550 lr=5.458000e-06 elapsed=01:00:22 avg_step=1.33s recent_step=1.54s eta=35:51:25 +2026-08-25 04:56:04 | INFO | train rank=0 | step=2740/100000 progress=2.7% loss=0.138550 lr=5.478000e-06 elapsed=01:00:34 avg_step=1.33s recent_step=1.18s eta=35:50:18 +2026-08-25 04:56:17 | INFO | train rank=0 | step=2750/100000 progress=2.8% loss=0.135376 lr=5.498000e-06 elapsed=01:00:47 avg_step=1.33s recent_step=1.23s eta=35:49:31 +2026-08-25 04:56:30 | INFO | train rank=0 | step=2760/100000 progress=2.8% loss=0.099487 lr=5.518000e-06 elapsed=01:01:00 avg_step=1.33s recent_step=1.35s eta=35:49:25 +2026-08-25 04:56:45 | INFO | train rank=0 | step=2770/100000 progress=2.8% loss=0.133545 lr=5.538000e-06 elapsed=01:01:15 avg_step=1.33s recent_step=1.54s eta=35:50:27 +2026-08-25 04:56:58 | INFO | train rank=0 | step=2780/100000 progress=2.8% loss=0.113586 lr=5.558000e-06 elapsed=01:01:27 avg_step=1.33s recent_step=1.20s eta=35:49:30 +2026-08-25 04:57:10 | INFO | train rank=0 | step=2790/100000 progress=2.8% loss=0.122559 lr=5.578000e-06 elapsed=01:01:39 avg_step=1.33s recent_step=1.20s eta=35:48:33 +2026-08-25 04:57:23 | INFO | train rank=0 | step=2800/100000 progress=2.8% loss=0.085510 lr=5.598000e-06 elapsed=01:01:53 avg_step=1.33s recent_step=1.35s eta=35:48:29 +2026-08-25 04:57:36 | INFO | train rank=0 | step=2810/100000 progress=2.8% loss=0.138672 lr=5.618000e-06 elapsed=01:02:06 avg_step=1.33s recent_step=1.27s eta=35:47:57 +2026-08-25 04:57:48 | INFO | train rank=0 | step=2820/100000 progress=2.8% loss=0.126221 lr=5.638000e-06 elapsed=01:02:18 avg_step=1.33s recent_step=1.25s eta=35:47:18 +2026-08-25 04:58:00 | INFO | train rank=0 | step=2830/100000 progress=2.8% loss=0.125977 lr=5.658000e-06 elapsed=01:02:30 avg_step=1.33s recent_step=1.20s eta=35:46:22 +2026-08-25 04:58:13 | INFO | train rank=0 | step=2840/100000 progress=2.8% loss=0.116455 lr=5.678000e-06 elapsed=01:02:43 avg_step=1.33s recent_step=1.25s eta=35:45:44 +2026-08-25 04:58:25 | INFO | train rank=0 | step=2850/100000 progress=2.9% loss=0.160522 lr=5.698000e-06 elapsed=01:02:55 avg_step=1.32s recent_step=1.25s eta=35:45:05 +2026-08-25 04:58:40 | INFO | train rank=0 | step=2860/100000 progress=2.9% loss=0.122925 lr=5.718000e-06 elapsed=01:03:10 avg_step=1.33s recent_step=1.44s eta=35:45:32 +2026-08-25 04:58:55 | INFO | train rank=0 | step=2870/100000 progress=2.9% loss=0.108521 lr=5.738000e-06 elapsed=01:03:24 avg_step=1.33s recent_step=1.48s eta=35:46:10 +2026-08-25 04:59:07 | INFO | train rank=0 | step=2880/100000 progress=2.9% loss=0.183472 lr=5.758000e-06 elapsed=01:03:37 avg_step=1.33s recent_step=1.25s eta=35:45:31 +2026-08-25 04:59:21 | INFO | train rank=0 | step=2890/100000 progress=2.9% loss=0.074005 lr=5.778000e-06 elapsed=01:03:51 avg_step=1.33s recent_step=1.44s eta=35:45:55 +2026-08-25 04:59:33 | INFO | train rank=0 | step=2900/100000 progress=2.9% loss=0.148987 lr=5.798000e-06 elapsed=01:04:03 avg_step=1.33s recent_step=1.20s eta=35:45:01 +2026-08-25 04:59:46 | INFO | train rank=0 | step=2910/100000 progress=2.9% loss=0.144958 lr=5.818000e-06 elapsed=01:04:16 avg_step=1.33s recent_step=1.29s eta=35:44:35 +2026-08-25 04:59:58 | INFO | train rank=0 | step=2920/100000 progress=2.9% loss=0.125122 lr=5.838000e-06 elapsed=01:04:28 avg_step=1.32s recent_step=1.20s eta=35:43:39 +2026-08-25 05:00:15 | INFO | train rank=0 | step=2930/100000 progress=2.9% loss=0.097290 lr=5.858000e-06 elapsed=01:04:45 avg_step=1.33s recent_step=1.72s eta=35:45:37 +2026-08-25 05:00:27 | INFO | train rank=0 | step=2940/100000 progress=2.9% loss=0.132690 lr=5.878000e-06 elapsed=01:04:57 avg_step=1.33s recent_step=1.18s eta=35:44:37 +2026-08-25 05:00:40 | INFO | train rank=0 | step=2950/100000 progress=2.9% loss=0.097473 lr=5.898000e-06 elapsed=01:05:09 avg_step=1.33s recent_step=1.22s eta=35:43:49 +2026-08-25 05:00:55 | INFO | train rank=0 | step=2960/100000 progress=3.0% loss=0.121887 lr=5.918000e-06 elapsed=01:05:25 avg_step=1.33s recent_step=1.51s eta=35:44:36 +2026-08-25 05:01:09 | INFO | train rank=0 | step=2970/100000 progress=3.0% loss=0.150452 lr=5.938000e-06 elapsed=01:05:39 avg_step=1.33s recent_step=1.41s eta=35:44:50 +2026-08-25 05:01:21 | INFO | train rank=0 | step=2980/100000 progress=3.0% loss=0.145508 lr=5.958000e-06 elapsed=01:05:51 avg_step=1.33s recent_step=1.25s eta=35:44:13 +2026-08-25 05:01:36 | INFO | train rank=0 | step=2990/100000 progress=3.0% loss=0.106537 lr=5.978000e-06 elapsed=01:06:06 avg_step=1.33s recent_step=1.49s eta=35:44:52 +2026-08-25 05:01:48 | INFO | train rank=0 | step=3000/100000 progress=3.0% loss=0.133911 lr=5.998000e-06 elapsed=01:06:18 avg_step=1.33s recent_step=1.16s eta=35:43:44 +2026-08-25 05:02:02 | INFO | train rank=0 | step=3010/100000 progress=3.0% loss=0.074860 lr=6.018000e-06 elapsed=01:06:32 avg_step=1.33s recent_step=1.41s eta=35:43:59 +2026-08-25 05:02:13 | INFO | train rank=0 | step=3020/100000 progress=3.0% loss=0.116638 lr=6.038000e-06 elapsed=01:06:43 avg_step=1.33s recent_step=1.12s eta=35:42:40 +2026-08-25 05:02:32 | INFO | train rank=0 | step=3030/100000 progress=3.0% loss=0.139893 lr=6.058000e-06 elapsed=01:07:01 avg_step=1.33s recent_step=1.86s eta=35:45:17 +2026-08-25 05:02:43 | INFO | train rank=0 | step=3040/100000 progress=3.0% loss=0.105743 lr=6.078000e-06 elapsed=01:07:13 avg_step=1.33s recent_step=1.18s eta=35:44:17 +2026-08-25 05:02:55 | INFO | train rank=0 | step=3050/100000 progress=3.0% loss=0.085693 lr=6.098000e-06 elapsed=01:07:25 avg_step=1.33s recent_step=1.19s eta=35:43:21 +2026-08-25 05:03:07 | INFO | train rank=0 | step=3060/100000 progress=3.1% loss=0.220093 lr=6.118000e-06 elapsed=01:07:37 avg_step=1.33s recent_step=1.21s eta=35:42:29 +2026-08-25 05:03:21 | INFO | train rank=0 | step=3070/100000 progress=3.1% loss=0.122437 lr=6.138000e-06 elapsed=01:07:50 avg_step=1.33s recent_step=1.32s eta=35:42:14 +2026-08-25 05:03:32 | INFO | train rank=0 | step=3080/100000 progress=3.1% loss=0.128540 lr=6.158000e-06 elapsed=01:08:02 avg_step=1.33s recent_step=1.17s eta=35:41:12 +2026-08-25 05:03:46 | INFO | train rank=0 | step=3090/100000 progress=3.1% loss=0.132568 lr=6.178000e-06 elapsed=01:08:16 avg_step=1.33s recent_step=1.40s eta=35:41:21 +2026-08-25 05:04:00 | INFO | train rank=0 | step=3100/100000 progress=3.1% loss=0.147278 lr=6.198000e-06 elapsed=01:08:30 avg_step=1.33s recent_step=1.35s eta=35:41:15 +2026-08-25 05:04:15 | INFO | train rank=0 | step=3110/100000 progress=3.1% loss=0.074493 lr=6.218000e-06 elapsed=01:08:45 avg_step=1.33s recent_step=1.51s eta=35:41:58 +2026-08-25 05:04:31 | INFO | train rank=0 | step=3120/100000 progress=3.1% loss=0.125488 lr=6.238000e-06 elapsed=01:09:01 avg_step=1.33s recent_step=1.60s eta=35:43:11 +2026-08-25 05:04:44 | INFO | train rank=0 | step=3130/100000 progress=3.1% loss=0.132629 lr=6.258000e-06 elapsed=01:09:14 avg_step=1.33s recent_step=1.32s eta=35:42:55 +2026-08-25 05:04:56 | INFO | train rank=0 | step=3140/100000 progress=3.1% loss=0.122620 lr=6.278000e-06 elapsed=01:09:26 avg_step=1.33s recent_step=1.23s eta=35:42:12 +2026-08-25 05:05:12 | INFO | train rank=0 | step=3150/100000 progress=3.1% loss=0.121948 lr=6.298000e-06 elapsed=01:09:42 avg_step=1.33s recent_step=1.54s eta=35:43:05 +2026-08-25 05:05:23 | INFO | train rank=0 | step=3160/100000 progress=3.2% loss=0.175415 lr=6.318000e-06 elapsed=01:09:53 avg_step=1.33s recent_step=1.16s eta=35:42:01 +2026-08-25 05:05:36 | INFO | train rank=0 | step=3170/100000 progress=3.2% loss=0.182617 lr=6.338000e-06 elapsed=01:10:06 avg_step=1.33s recent_step=1.24s eta=35:41:20 +2026-08-25 05:05:48 | INFO | train rank=0 | step=3180/100000 progress=3.2% loss=0.138733 lr=6.358000e-06 elapsed=01:10:18 avg_step=1.33s recent_step=1.19s eta=35:40:26 +2026-08-25 05:06:02 | INFO | train rank=0 | step=3190/100000 progress=3.2% loss=0.090210 lr=6.378000e-06 elapsed=01:10:32 avg_step=1.33s recent_step=1.44s eta=35:40:47 +2026-08-25 05:06:14 | INFO | train rank=0 | step=3200/100000 progress=3.2% loss=0.134094 lr=6.398000e-06 elapsed=01:10:44 avg_step=1.33s recent_step=1.20s eta=35:39:54 +2026-08-25 05:06:25 | INFO | train rank=0 | step=3210/100000 progress=3.2% loss=0.116272 lr=6.418000e-06 elapsed=01:10:55 avg_step=1.33s recent_step=1.14s eta=35:38:44 +2026-08-25 05:06:38 | INFO | train rank=0 | step=3220/100000 progress=3.2% loss=0.102905 lr=6.438000e-06 elapsed=01:11:08 avg_step=1.33s recent_step=1.22s eta=35:37:59 +2026-08-25 05:06:49 | INFO | train rank=0 | step=3230/100000 progress=3.2% loss=0.082245 lr=6.458000e-06 elapsed=01:11:19 avg_step=1.32s recent_step=1.11s eta=35:36:41 +2026-08-25 05:07:01 | INFO | train rank=0 | step=3240/100000 progress=3.2% loss=0.097168 lr=6.478000e-06 elapsed=01:11:31 avg_step=1.32s recent_step=1.24s eta=35:36:02 +2026-08-25 05:07:15 | INFO | train rank=0 | step=3250/100000 progress=3.2% loss=0.173462 lr=6.498000e-06 elapsed=01:11:45 avg_step=1.32s recent_step=1.41s eta=35:36:14 +2026-08-25 05:07:27 | INFO | train rank=0 | step=3260/100000 progress=3.3% loss=0.092896 lr=6.518000e-06 elapsed=01:11:57 avg_step=1.32s recent_step=1.15s eta=35:35:10 +2026-08-25 05:07:44 | INFO | train rank=0 | step=3270/100000 progress=3.3% loss=0.108765 lr=6.538000e-06 elapsed=01:12:14 avg_step=1.33s recent_step=1.69s eta=35:36:46 +2026-08-25 05:08:00 | INFO | train rank=0 | step=3280/100000 progress=3.3% loss=0.117218 lr=6.558000e-06 elapsed=01:12:30 avg_step=1.33s recent_step=1.61s eta=35:37:56 +2026-08-25 05:08:11 | INFO | train rank=0 | step=3290/100000 progress=3.3% loss=0.143921 lr=6.578000e-06 elapsed=01:12:41 avg_step=1.33s recent_step=1.16s eta=35:36:53 +2026-08-25 05:08:23 | INFO | train rank=0 | step=3300/100000 progress=3.3% loss=0.116638 lr=6.598000e-06 elapsed=01:12:53 avg_step=1.33s recent_step=1.19s eta=35:36:02 +2026-08-25 05:08:35 | INFO | train rank=0 | step=3310/100000 progress=3.3% loss=0.081329 lr=6.618000e-06 elapsed=01:13:05 avg_step=1.32s recent_step=1.17s eta=35:35:04 +2026-08-25 05:08:47 | INFO | train rank=0 | step=3320/100000 progress=3.3% loss=0.128906 lr=6.638000e-06 elapsed=01:13:17 avg_step=1.32s recent_step=1.20s eta=35:34:16 +2026-08-25 05:08:59 | INFO | train rank=0 | step=3330/100000 progress=3.3% loss=0.111572 lr=6.658000e-06 elapsed=01:13:29 avg_step=1.32s recent_step=1.21s eta=35:33:31 +2026-08-25 05:09:11 | INFO | train rank=0 | step=3340/100000 progress=3.3% loss=0.102173 lr=6.678000e-06 elapsed=01:13:41 avg_step=1.32s recent_step=1.16s eta=35:32:31 +2026-08-25 05:09:29 | INFO | train rank=0 | step=3350/100000 progress=3.4% loss=0.081604 lr=6.698000e-06 elapsed=01:13:59 avg_step=1.33s recent_step=1.83s eta=35:34:45 +2026-08-25 05:09:41 | INFO | train rank=0 | step=3360/100000 progress=3.4% loss=0.195801 lr=6.718000e-06 elapsed=01:14:11 avg_step=1.32s recent_step=1.16s eta=35:33:46 +2026-08-25 05:09:53 | INFO | train rank=0 | step=3370/100000 progress=3.4% loss=0.140625 lr=6.738000e-06 elapsed=01:14:23 avg_step=1.32s recent_step=1.20s eta=35:32:56 +2026-08-25 05:10:06 | INFO | train rank=0 | step=3380/100000 progress=3.4% loss=0.134949 lr=6.758000e-06 elapsed=01:14:35 avg_step=1.32s recent_step=1.27s eta=35:32:27 +2026-08-25 05:10:23 | INFO | train rank=0 | step=3390/100000 progress=3.4% loss=0.133301 lr=6.778000e-06 elapsed=01:14:53 avg_step=1.33s recent_step=1.74s eta=35:34:12 +2026-08-25 05:10:37 | INFO | train rank=0 | step=3400/100000 progress=3.4% loss=0.111816 lr=6.798000e-06 elapsed=01:15:07 avg_step=1.33s recent_step=1.44s eta=35:34:32 +2026-08-25 05:10:50 | INFO | train rank=0 | step=3410/100000 progress=3.4% loss=0.116974 lr=6.818000e-06 elapsed=01:15:20 avg_step=1.33s recent_step=1.25s eta=35:33:57 +2026-08-25 05:11:01 | INFO | train rank=0 | step=3420/100000 progress=3.4% loss=0.088470 lr=6.838000e-06 elapsed=01:15:31 avg_step=1.33s recent_step=1.14s eta=35:32:50 +2026-08-25 05:11:14 | INFO | train rank=0 | step=3430/100000 progress=3.4% loss=0.113892 lr=6.858000e-06 elapsed=01:15:44 avg_step=1.32s recent_step=1.24s eta=35:32:14 +2026-08-25 05:11:29 | INFO | train rank=0 | step=3440/100000 progress=3.4% loss=0.169067 lr=6.878000e-06 elapsed=01:15:58 avg_step=1.33s recent_step=1.49s eta=35:32:48 +2026-08-25 05:11:40 | INFO | train rank=0 | step=3450/100000 progress=3.5% loss=0.127930 lr=6.898000e-06 elapsed=01:16:10 avg_step=1.32s recent_step=1.18s eta=35:31:54 +2026-08-25 05:11:58 | INFO | train rank=0 | step=3460/100000 progress=3.5% loss=0.115784 lr=6.918000e-06 elapsed=01:16:28 avg_step=1.33s recent_step=1.77s eta=35:33:46 +2026-08-25 05:12:10 | INFO | train rank=0 | step=3470/100000 progress=3.5% loss=0.114319 lr=6.938000e-06 elapsed=01:16:40 avg_step=1.33s recent_step=1.16s eta=35:32:48 +2026-08-25 05:12:22 | INFO | train rank=0 | step=3480/100000 progress=3.5% loss=0.182129 lr=6.958000e-06 elapsed=01:16:52 avg_step=1.33s recent_step=1.20s eta=35:32:01 +2026-08-25 05:12:34 | INFO | train rank=0 | step=3490/100000 progress=3.5% loss=0.124451 lr=6.978000e-06 elapsed=01:17:04 avg_step=1.33s recent_step=1.22s eta=35:31:17 +2026-08-25 05:12:49 | INFO | train rank=0 | step=3500/100000 progress=3.5% loss=0.099792 lr=6.998000e-06 elapsed=01:17:19 avg_step=1.33s recent_step=1.50s eta=35:31:53 +2026-08-25 05:13:01 | INFO | train rank=0 | step=3510/100000 progress=3.5% loss=0.104431 lr=7.018000e-06 elapsed=01:17:31 avg_step=1.33s recent_step=1.23s eta=35:31:13 +2026-08-25 05:13:13 | INFO | train rank=0 | step=3520/100000 progress=3.5% loss=0.082153 lr=7.038000e-06 elapsed=01:17:43 avg_step=1.32s recent_step=1.18s eta=35:30:20 +2026-08-25 05:13:25 | INFO | train rank=0 | step=3530/100000 progress=3.5% loss=0.117371 lr=7.058000e-06 elapsed=01:17:55 avg_step=1.32s recent_step=1.17s eta=35:29:25 +2026-08-25 05:13:37 | INFO | train rank=0 | step=3540/100000 progress=3.5% loss=0.093018 lr=7.078000e-06 elapsed=01:18:07 avg_step=1.32s recent_step=1.24s eta=35:28:47 +2026-08-25 05:13:51 | INFO | train rank=0 | step=3550/100000 progress=3.5% loss=0.137512 lr=7.098000e-06 elapsed=01:18:21 avg_step=1.32s recent_step=1.44s eta=35:29:04 +2026-08-25 05:14:04 | INFO | train rank=0 | step=3560/100000 progress=3.6% loss=0.084961 lr=7.118000e-06 elapsed=01:18:33 avg_step=1.32s recent_step=1.21s eta=35:28:20 +2026-08-25 05:14:18 | INFO | train rank=0 | step=3570/100000 progress=3.6% loss=0.133545 lr=7.138000e-06 elapsed=01:18:48 avg_step=1.32s recent_step=1.41s eta=35:28:28 +2026-08-25 05:14:29 | INFO | train rank=0 | step=3580/100000 progress=3.6% loss=0.103333 lr=7.158000e-06 elapsed=01:18:59 avg_step=1.32s recent_step=1.18s eta=35:27:36 +2026-08-25 05:14:46 | INFO | train rank=0 | step=3590/100000 progress=3.6% loss=0.128845 lr=7.178000e-06 elapsed=01:19:16 avg_step=1.33s recent_step=1.70s eta=35:29:04 +2026-08-25 05:15:00 | INFO | train rank=0 | step=3600/100000 progress=3.6% loss=0.103394 lr=7.198000e-06 elapsed=01:19:30 avg_step=1.33s recent_step=1.33s eta=35:28:53 +2026-08-25 05:15:11 | INFO | train rank=0 | step=3610/100000 progress=3.6% loss=0.086243 lr=7.218000e-06 elapsed=01:19:41 avg_step=1.32s recent_step=1.14s eta=35:27:51 +2026-08-25 05:15:26 | INFO | train rank=0 | step=3620/100000 progress=3.6% loss=0.117737 lr=7.238000e-06 elapsed=01:19:56 avg_step=1.32s recent_step=1.45s eta=35:28:11 +2026-08-25 05:15:37 | INFO | train rank=0 | step=3630/100000 progress=3.6% loss=0.076813 lr=7.258000e-06 elapsed=01:20:07 avg_step=1.32s recent_step=1.17s eta=35:27:18 +2026-08-25 05:15:52 | INFO | train rank=0 | step=3640/100000 progress=3.6% loss=0.082977 lr=7.278000e-06 elapsed=01:20:22 avg_step=1.32s recent_step=1.48s eta=35:27:47 +2026-08-25 05:16:04 | INFO | train rank=0 | step=3650/100000 progress=3.6% loss=0.102203 lr=7.298000e-06 elapsed=01:20:34 avg_step=1.32s recent_step=1.20s eta=35:26:59 +2026-08-25 05:16:16 | INFO | train rank=0 | step=3660/100000 progress=3.7% loss=0.131714 lr=7.318000e-06 elapsed=01:20:46 avg_step=1.32s recent_step=1.17s eta=35:26:06 +2026-08-25 05:16:32 | INFO | train rank=0 | step=3670/100000 progress=3.7% loss=0.091034 lr=7.338000e-06 elapsed=01:21:01 avg_step=1.32s recent_step=1.56s eta=35:26:55 +2026-08-25 05:16:44 | INFO | train rank=0 | step=3680/100000 progress=3.7% loss=0.067200 lr=7.358000e-06 elapsed=01:21:14 avg_step=1.32s recent_step=1.24s eta=35:26:20 +2026-08-25 05:16:56 | INFO | train rank=0 | step=3690/100000 progress=3.7% loss=0.104919 lr=7.378000e-06 elapsed=01:21:26 avg_step=1.32s recent_step=1.24s eta=35:25:43 +2026-08-25 05:17:10 | INFO | train rank=0 | step=3700/100000 progress=3.7% loss=0.131897 lr=7.398000e-06 elapsed=01:21:40 avg_step=1.32s recent_step=1.34s eta=35:25:35 +2026-08-25 05:17:22 | INFO | train rank=0 | step=3710/100000 progress=3.7% loss=0.161560 lr=7.418000e-06 elapsed=01:21:52 avg_step=1.32s recent_step=1.26s eta=35:25:04 +2026-08-25 05:17:37 | INFO | train rank=0 | step=3720/100000 progress=3.7% loss=0.098053 lr=7.438000e-06 elapsed=01:22:07 avg_step=1.32s recent_step=1.51s eta=35:25:39 +2026-08-25 05:17:50 | INFO | train rank=0 | step=3730/100000 progress=3.7% loss=0.161621 lr=7.458000e-06 elapsed=01:22:19 avg_step=1.32s recent_step=1.22s eta=35:24:58 +2026-08-25 05:18:02 | INFO | train rank=0 | step=3740/100000 progress=3.7% loss=0.109192 lr=7.478000e-06 elapsed=01:22:32 avg_step=1.32s recent_step=1.22s eta=35:24:17 +2026-08-25 05:18:17 | INFO | train rank=0 | step=3750/100000 progress=3.8% loss=0.089844 lr=7.498000e-06 elapsed=01:22:47 avg_step=1.32s recent_step=1.51s eta=35:24:51 +2026-08-25 05:18:28 | INFO | train rank=0 | step=3760/100000 progress=3.8% loss=0.141785 lr=7.518000e-06 elapsed=01:22:58 avg_step=1.32s recent_step=1.16s eta=35:23:55 +2026-08-25 05:18:41 | INFO | train rank=0 | step=3770/100000 progress=3.8% loss=0.104950 lr=7.538000e-06 elapsed=01:23:10 avg_step=1.32s recent_step=1.22s eta=35:23:15 +2026-08-25 05:18:55 | INFO | train rank=0 | step=3780/100000 progress=3.8% loss=0.097900 lr=7.558000e-06 elapsed=01:23:25 avg_step=1.32s recent_step=1.46s eta=35:23:37 +2026-08-25 05:19:12 | INFO | train rank=0 | step=3790/100000 progress=3.8% loss=0.133728 lr=7.578000e-06 elapsed=01:23:41 avg_step=1.33s recent_step=1.63s eta=35:24:42 +2026-08-25 05:19:24 | INFO | train rank=0 | step=3800/100000 progress=3.8% loss=0.145935 lr=7.598000e-06 elapsed=01:23:54 avg_step=1.32s recent_step=1.25s eta=35:24:09 +2026-08-25 05:19:36 | INFO | train rank=0 | step=3810/100000 progress=3.8% loss=0.128082 lr=7.618000e-06 elapsed=01:24:06 avg_step=1.32s recent_step=1.17s eta=35:23:16 +2026-08-25 05:19:50 | INFO | train rank=0 | step=3820/100000 progress=3.8% loss=0.152466 lr=7.638000e-06 elapsed=01:24:19 avg_step=1.32s recent_step=1.38s eta=35:23:18 +2026-08-25 05:20:06 | INFO | train rank=0 | step=3830/100000 progress=3.8% loss=0.103577 lr=7.658000e-06 elapsed=01:24:36 avg_step=1.33s recent_step=1.67s eta=35:24:32 +2026-08-25 05:20:18 | INFO | train rank=0 | step=3840/100000 progress=3.8% loss=0.139343 lr=7.678000e-06 elapsed=01:24:47 avg_step=1.32s recent_step=1.13s eta=35:23:31 +2026-08-25 05:20:31 | INFO | train rank=0 | step=3850/100000 progress=3.9% loss=0.097900 lr=7.698000e-06 elapsed=01:25:01 avg_step=1.33s recent_step=1.39s eta=35:23:33 +2026-08-25 05:20:46 | INFO | train rank=0 | step=3860/100000 progress=3.9% loss=0.152588 lr=7.718000e-06 elapsed=01:25:16 avg_step=1.33s recent_step=1.48s eta=35:23:59 +2026-08-25 05:20:58 | INFO | train rank=0 | step=3870/100000 progress=3.9% loss=0.074249 lr=7.738000e-06 elapsed=01:25:28 avg_step=1.33s recent_step=1.14s eta=35:23:00 +2026-08-25 05:21:10 | INFO | train rank=0 | step=3880/100000 progress=3.9% loss=0.079834 lr=7.758000e-06 elapsed=01:25:40 avg_step=1.32s recent_step=1.20s eta=35:22:17 +2026-08-25 05:21:24 | INFO | train rank=0 | step=3890/100000 progress=3.9% loss=0.125671 lr=7.778000e-06 elapsed=01:25:54 avg_step=1.33s recent_step=1.44s eta=35:22:32 +2026-08-25 05:21:36 | INFO | train rank=0 | step=3900/100000 progress=3.9% loss=0.093048 lr=7.798000e-06 elapsed=01:26:06 avg_step=1.32s recent_step=1.16s eta=35:21:38 +2026-08-25 05:21:50 | INFO | train rank=0 | step=3910/100000 progress=3.9% loss=0.086060 lr=7.818000e-06 elapsed=01:26:20 avg_step=1.32s recent_step=1.46s eta=35:21:57 +2026-08-25 05:22:02 | INFO | train rank=0 | step=3920/100000 progress=3.9% loss=0.080048 lr=7.838000e-06 elapsed=01:26:32 avg_step=1.32s recent_step=1.14s eta=35:20:58 +2026-08-25 05:22:14 | INFO | train rank=0 | step=3930/100000 progress=3.9% loss=0.069794 lr=7.858000e-06 elapsed=01:26:44 avg_step=1.32s recent_step=1.24s eta=35:20:25 +2026-08-25 05:22:26 | INFO | train rank=0 | step=3940/100000 progress=3.9% loss=0.146362 lr=7.878000e-06 elapsed=01:26:56 avg_step=1.32s recent_step=1.22s eta=35:19:47 +2026-08-25 05:22:39 | INFO | train rank=0 | step=3950/100000 progress=4.0% loss=0.117523 lr=7.898000e-06 elapsed=01:27:08 avg_step=1.32s recent_step=1.22s eta=35:19:08 +2026-08-25 05:22:53 | INFO | train rank=0 | step=3960/100000 progress=4.0% loss=0.093750 lr=7.918000e-06 elapsed=01:27:23 avg_step=1.32s recent_step=1.46s eta=35:19:28 +2026-08-25 05:23:07 | INFO | train rank=0 | step=3970/100000 progress=4.0% loss=0.076721 lr=7.938000e-06 elapsed=01:27:37 avg_step=1.32s recent_step=1.39s eta=35:19:31 +2026-08-25 05:23:22 | INFO | train rank=0 | step=3980/100000 progress=4.0% loss=0.131226 lr=7.958000e-06 elapsed=01:27:52 avg_step=1.32s recent_step=1.46s eta=35:19:52 +2026-08-25 05:23:40 | INFO | train rank=0 | step=3990/100000 progress=4.0% loss=0.122498 lr=7.978000e-06 elapsed=01:28:10 avg_step=1.33s recent_step=1.82s eta=35:21:37 +2026-08-25 05:23:53 | INFO | train rank=0 | step=4000/100000 progress=4.0% loss=0.129639 lr=7.998000e-06 elapsed=01:28:23 avg_step=1.33s recent_step=1.32s eta=35:21:21 +2026-08-25 05:24:05 | INFO | train rank=0 | step=4010/100000 progress=4.0% loss=0.058594 lr=8.018000e-06 elapsed=01:28:35 avg_step=1.33s recent_step=1.18s eta=35:20:33 +2026-08-25 05:24:17 | INFO | train rank=0 | step=4020/100000 progress=4.0% loss=0.088043 lr=8.038000e-06 elapsed=01:28:47 avg_step=1.33s recent_step=1.21s eta=35:19:53 +2026-08-25 05:24:32 | INFO | train rank=0 | step=4030/100000 progress=4.0% loss=0.105042 lr=8.058000e-06 elapsed=01:29:02 avg_step=1.33s recent_step=1.47s eta=35:20:14 +2026-08-25 05:24:44 | INFO | train rank=0 | step=4040/100000 progress=4.0% loss=0.134033 lr=8.078000e-06 elapsed=01:29:14 avg_step=1.33s recent_step=1.20s eta=35:19:32 +2026-08-25 05:24:55 | INFO | train rank=0 | step=4050/100000 progress=4.0% loss=0.145630 lr=8.098000e-06 elapsed=01:29:25 avg_step=1.32s recent_step=1.15s eta=35:18:36 +2026-08-25 05:25:07 | INFO | train rank=0 | step=4060/100000 progress=4.1% loss=0.111450 lr=8.118000e-06 elapsed=01:29:37 avg_step=1.32s recent_step=1.20s eta=35:17:53 +2026-08-25 05:25:20 | INFO | train rank=0 | step=4070/100000 progress=4.1% loss=0.073975 lr=8.138000e-06 elapsed=01:29:50 avg_step=1.32s recent_step=1.33s eta=35:17:40 +2026-08-25 05:25:34 | INFO | train rank=0 | step=4080/100000 progress=4.1% loss=0.088165 lr=8.158000e-06 elapsed=01:30:04 avg_step=1.32s recent_step=1.36s eta=35:17:35 +2026-08-25 05:25:46 | INFO | train rank=0 | step=4090/100000 progress=4.1% loss=0.166992 lr=8.178000e-06 elapsed=01:30:16 avg_step=1.32s recent_step=1.19s eta=35:16:51 +2026-08-25 05:25:58 | INFO | train rank=0 | step=4100/100000 progress=4.1% loss=0.073151 lr=8.198000e-06 elapsed=01:30:28 avg_step=1.32s recent_step=1.17s eta=35:16:02 +2026-08-25 05:26:15 | INFO | train rank=0 | step=4110/100000 progress=4.1% loss=0.108673 lr=8.218000e-06 elapsed=01:30:45 avg_step=1.32s recent_step=1.72s eta=35:17:22 +2026-08-25 05:26:28 | INFO | train rank=0 | step=4120/100000 progress=4.1% loss=0.108948 lr=8.238000e-06 elapsed=01:30:58 avg_step=1.32s recent_step=1.29s eta=35:17:01 +2026-08-25 05:26:40 | INFO | train rank=0 | step=4130/100000 progress=4.1% loss=0.118286 lr=8.258000e-06 elapsed=01:31:10 avg_step=1.32s recent_step=1.23s eta=35:16:24 +2026-08-25 05:26:54 | INFO | train rank=0 | step=4140/100000 progress=4.1% loss=0.091919 lr=8.278000e-06 elapsed=01:31:24 avg_step=1.32s recent_step=1.41s eta=35:16:31 +2026-08-25 05:27:11 | INFO | train rank=0 | step=4150/100000 progress=4.2% loss=0.086914 lr=8.298000e-06 elapsed=01:31:41 avg_step=1.33s recent_step=1.73s eta=35:17:51 +2026-08-25 05:27:23 | INFO | train rank=0 | step=4160/100000 progress=4.2% loss=0.107849 lr=8.318000e-06 elapsed=01:31:53 avg_step=1.33s recent_step=1.12s eta=35:16:51 +2026-08-25 05:27:35 | INFO | train rank=0 | step=4170/100000 progress=4.2% loss=0.145874 lr=8.338000e-06 elapsed=01:32:05 avg_step=1.33s recent_step=1.28s eta=35:16:28 +2026-08-25 05:27:49 | INFO | train rank=0 | step=4180/100000 progress=4.2% loss=0.062927 lr=8.358000e-06 elapsed=01:32:19 avg_step=1.33s recent_step=1.40s eta=35:16:32 +2026-08-25 05:28:01 | INFO | train rank=0 | step=4190/100000 progress=4.2% loss=0.103638 lr=8.378000e-06 elapsed=01:32:31 avg_step=1.32s recent_step=1.16s eta=35:15:42 +2026-08-25 05:28:13 | INFO | train rank=0 | step=4200/100000 progress=4.2% loss=0.096558 lr=8.398000e-06 elapsed=01:32:43 avg_step=1.32s recent_step=1.20s eta=35:15:00 +2026-08-25 05:28:25 | INFO | train rank=0 | step=4210/100000 progress=4.2% loss=0.079712 lr=8.418000e-06 elapsed=01:32:55 avg_step=1.32s recent_step=1.18s eta=35:14:13 +2026-08-25 05:28:38 | INFO | train rank=0 | step=4220/100000 progress=4.2% loss=0.082764 lr=8.438000e-06 elapsed=01:33:08 avg_step=1.32s recent_step=1.35s eta=35:14:05 +2026-08-25 05:28:52 | INFO | train rank=0 | step=4230/100000 progress=4.2% loss=0.129761 lr=8.458000e-06 elapsed=01:33:22 avg_step=1.32s recent_step=1.34s eta=35:13:54 +2026-08-25 05:29:03 | INFO | train rank=0 | step=4240/100000 progress=4.2% loss=0.094910 lr=8.478000e-06 elapsed=01:33:33 avg_step=1.32s recent_step=1.14s eta=35:13:00 +2026-08-25 05:29:17 | INFO | train rank=0 | step=4250/100000 progress=4.2% loss=0.182983 lr=8.498000e-06 elapsed=01:33:47 avg_step=1.32s recent_step=1.38s eta=35:12:58 +2026-08-25 05:29:29 | INFO | train rank=0 | step=4260/100000 progress=4.3% loss=0.093201 lr=8.518000e-06 elapsed=01:33:59 avg_step=1.32s recent_step=1.18s eta=35:12:13 +2026-08-25 05:29:48 | INFO | train rank=0 | step=4270/100000 progress=4.3% loss=0.087067 lr=8.538000e-06 elapsed=01:34:17 avg_step=1.33s recent_step=1.89s eta=35:14:06 +2026-08-25 05:29:59 | INFO | train rank=0 | step=4280/100000 progress=4.3% loss=0.099976 lr=8.558000e-06 elapsed=01:34:29 avg_step=1.32s recent_step=1.19s eta=35:13:22 +2026-08-25 05:30:15 | INFO | train rank=0 | step=4290/100000 progress=4.3% loss=0.088745 lr=8.578000e-06 elapsed=01:34:45 avg_step=1.33s recent_step=1.56s eta=35:14:02 +2026-08-25 05:30:29 | INFO | train rank=0 | step=4300/100000 progress=4.3% loss=0.231445 lr=8.598000e-06 elapsed=01:34:59 avg_step=1.33s recent_step=1.44s eta=35:14:13 +2026-08-25 05:30:41 | INFO | train rank=0 | step=4310/100000 progress=4.3% loss=0.095825 lr=8.618000e-06 elapsed=01:35:11 avg_step=1.33s recent_step=1.19s eta=35:13:29 +2026-08-25 05:30:53 | INFO | train rank=0 | step=4320/100000 progress=4.3% loss=0.093903 lr=8.638000e-06 elapsed=01:35:23 avg_step=1.32s recent_step=1.21s eta=35:12:51 +2026-08-25 05:31:11 | INFO | train rank=0 | step=4330/100000 progress=4.3% loss=0.077148 lr=8.658000e-06 elapsed=01:35:41 avg_step=1.33s recent_step=1.73s eta=35:14:08 +2026-08-25 05:31:22 | INFO | train rank=0 | step=4340/100000 progress=4.3% loss=0.125122 lr=8.678000e-06 elapsed=01:35:52 avg_step=1.33s recent_step=1.15s eta=35:13:15 +2026-08-25 05:31:34 | INFO | train rank=0 | step=4350/100000 progress=4.3% loss=0.141113 lr=8.698000e-06 elapsed=01:36:04 avg_step=1.33s recent_step=1.20s eta=35:12:33 +2026-08-25 05:31:46 | INFO | train rank=0 | step=4360/100000 progress=4.4% loss=0.144775 lr=8.718000e-06 elapsed=01:36:16 avg_step=1.32s recent_step=1.23s eta=35:11:58 +2026-08-25 05:31:58 | INFO | train rank=0 | step=4370/100000 progress=4.4% loss=0.076660 lr=8.738000e-06 elapsed=01:36:28 avg_step=1.32s recent_step=1.15s eta=35:11:06 +2026-08-25 05:32:10 | INFO | train rank=0 | step=4380/100000 progress=4.4% loss=0.079956 lr=8.758000e-06 elapsed=01:36:40 avg_step=1.32s recent_step=1.21s eta=35:10:29 +2026-08-25 05:32:22 | INFO | train rank=0 | step=4390/100000 progress=4.4% loss=0.086334 lr=8.778000e-06 elapsed=01:36:52 avg_step=1.32s recent_step=1.16s eta=35:09:40 +2026-08-25 05:32:33 | INFO | train rank=0 | step=4400/100000 progress=4.4% loss=0.083557 lr=8.798000e-06 elapsed=01:37:03 avg_step=1.32s recent_step=1.17s eta=35:08:52 +2026-08-25 05:32:45 | INFO | train rank=0 | step=4410/100000 progress=4.4% loss=0.105835 lr=8.818000e-06 elapsed=01:37:15 avg_step=1.32s recent_step=1.21s eta=35:08:15 +2026-08-25 05:32:58 | INFO | train rank=0 | step=4420/100000 progress=4.4% loss=0.118713 lr=8.838000e-06 elapsed=01:37:28 avg_step=1.32s recent_step=1.23s eta=35:07:43 +2026-08-25 05:33:12 | INFO | train rank=0 | step=4430/100000 progress=4.4% loss=0.077728 lr=8.858000e-06 elapsed=01:37:42 avg_step=1.32s recent_step=1.38s eta=35:07:43 +2026-08-25 05:33:23 | INFO | train rank=0 | step=4440/100000 progress=4.4% loss=0.118958 lr=8.878000e-06 elapsed=01:37:53 avg_step=1.32s recent_step=1.16s eta=35:06:54 +2026-08-25 05:33:34 | INFO | train rank=0 | step=4450/100000 progress=4.5% loss=0.088715 lr=8.898000e-06 elapsed=01:38:04 avg_step=1.32s recent_step=1.13s eta=35:05:59 +2026-08-25 05:33:49 | INFO | train rank=0 | step=4460/100000 progress=4.5% loss=0.087128 lr=8.918000e-06 elapsed=01:38:19 avg_step=1.32s recent_step=1.46s eta=35:06:17 +2026-08-25 05:34:02 | INFO | train rank=0 | step=4470/100000 progress=4.5% loss=0.115662 lr=8.938000e-06 elapsed=01:38:32 avg_step=1.32s recent_step=1.26s eta=35:05:50 +2026-08-25 05:34:14 | INFO | train rank=0 | step=4480/100000 progress=4.5% loss=0.188477 lr=8.958000e-06 elapsed=01:38:44 avg_step=1.32s recent_step=1.27s eta=35:05:26 +2026-08-25 05:34:26 | INFO | train rank=0 | step=4490/100000 progress=4.5% loss=0.062592 lr=8.978000e-06 elapsed=01:38:56 avg_step=1.32s recent_step=1.19s eta=35:04:44 +2026-08-25 05:34:43 | INFO | train rank=0 | step=4500/100000 progress=4.5% loss=0.066406 lr=8.998000e-06 elapsed=01:39:13 avg_step=1.32s recent_step=1.68s eta=35:05:47 +2026-08-25 05:34:56 | INFO | train rank=0 | step=4510/100000 progress=4.5% loss=0.150391 lr=9.018000e-06 elapsed=01:39:26 avg_step=1.32s recent_step=1.31s eta=35:05:31 +2026-08-25 05:35:08 | INFO | train rank=0 | step=4520/100000 progress=4.5% loss=0.130127 lr=9.038000e-06 elapsed=01:39:38 avg_step=1.32s recent_step=1.20s eta=35:04:51 +2026-08-25 05:35:21 | INFO | train rank=0 | step=4530/100000 progress=4.5% loss=0.071411 lr=9.058000e-06 elapsed=01:39:51 avg_step=1.32s recent_step=1.32s eta=35:04:36 +2026-08-25 05:35:37 | INFO | train rank=0 | step=4540/100000 progress=4.5% loss=0.103027 lr=9.078000e-06 elapsed=01:40:07 avg_step=1.32s recent_step=1.56s eta=35:05:12 +2026-08-25 05:35:51 | INFO | train rank=0 | step=4550/100000 progress=4.5% loss=0.161682 lr=9.098000e-06 elapsed=01:40:21 avg_step=1.32s recent_step=1.41s eta=35:05:17 +2026-08-25 05:36:07 | INFO | train rank=0 | step=4560/100000 progress=4.6% loss=0.101593 lr=9.118000e-06 elapsed=01:40:37 avg_step=1.32s recent_step=1.63s eta=35:06:09 +2026-08-25 05:36:20 | INFO | train rank=0 | step=4570/100000 progress=4.6% loss=0.226074 lr=9.138000e-06 elapsed=01:40:50 avg_step=1.32s recent_step=1.24s eta=35:05:39 +2026-08-25 05:36:35 | INFO | train rank=0 | step=4580/100000 progress=4.6% loss=0.129700 lr=9.158000e-06 elapsed=01:41:05 avg_step=1.32s recent_step=1.52s eta=35:06:07 +2026-08-25 05:36:48 | INFO | train rank=0 | step=4590/100000 progress=4.6% loss=0.087921 lr=9.178000e-06 elapsed=01:41:18 avg_step=1.32s recent_step=1.29s eta=35:05:47 +2026-08-25 05:37:05 | INFO | train rank=0 | step=4600/100000 progress=4.6% loss=0.080597 lr=9.198000e-06 elapsed=01:41:34 avg_step=1.32s recent_step=1.66s eta=35:06:43 +2026-08-25 05:37:16 | INFO | train rank=0 | step=4610/100000 progress=4.6% loss=0.061859 lr=9.218000e-06 elapsed=01:41:46 avg_step=1.32s recent_step=1.13s eta=35:05:49 +2026-08-25 05:37:29 | INFO | train rank=0 | step=4620/100000 progress=4.6% loss=0.081116 lr=9.238000e-06 elapsed=01:41:59 avg_step=1.32s recent_step=1.33s eta=35:05:36 +2026-08-25 05:37:42 | INFO | train rank=0 | step=4630/100000 progress=4.6% loss=0.120667 lr=9.258000e-06 elapsed=01:42:12 avg_step=1.32s recent_step=1.26s eta=35:05:11 +2026-08-25 05:37:54 | INFO | train rank=0 | step=4640/100000 progress=4.6% loss=0.082672 lr=9.278000e-06 elapsed=01:42:24 avg_step=1.32s recent_step=1.27s eta=35:04:45 +2026-08-25 05:38:08 | INFO | train rank=0 | step=4650/100000 progress=4.7% loss=0.083313 lr=9.298000e-06 elapsed=01:42:38 avg_step=1.32s recent_step=1.35s eta=35:04:38 +2026-08-25 05:38:20 | INFO | train rank=0 | step=4660/100000 progress=4.7% loss=0.083740 lr=9.318000e-06 elapsed=01:42:50 avg_step=1.32s recent_step=1.21s eta=35:04:00 +2026-08-25 05:38:33 | INFO | train rank=0 | step=4670/100000 progress=4.7% loss=0.048584 lr=9.338000e-06 elapsed=01:43:03 avg_step=1.32s recent_step=1.28s eta=35:03:38 +2026-08-25 05:38:45 | INFO | train rank=0 | step=4680/100000 progress=4.7% loss=0.088257 lr=9.358000e-06 elapsed=01:43:15 avg_step=1.32s recent_step=1.21s eta=35:03:01 +2026-08-25 05:38:57 | INFO | train rank=0 | step=4690/100000 progress=4.7% loss=0.111420 lr=9.378000e-06 elapsed=01:43:26 avg_step=1.32s recent_step=1.17s eta=35:02:16 +2026-08-25 05:39:09 | INFO | train rank=0 | step=4700/100000 progress=4.7% loss=0.105042 lr=9.398000e-06 elapsed=01:43:38 avg_step=1.32s recent_step=1.20s eta=35:01:39 +2026-08-25 05:39:24 | INFO | train rank=0 | step=4710/100000 progress=4.7% loss=0.075958 lr=9.418000e-06 elapsed=01:43:54 avg_step=1.32s recent_step=1.52s eta=35:02:06 +2026-08-25 05:39:36 | INFO | train rank=0 | step=4720/100000 progress=4.7% loss=0.075409 lr=9.438000e-06 elapsed=01:44:06 avg_step=1.32s recent_step=1.22s eta=35:01:31 +2026-08-25 05:39:48 | INFO | train rank=0 | step=4730/100000 progress=4.7% loss=0.118835 lr=9.458000e-06 elapsed=01:44:18 avg_step=1.32s recent_step=1.22s eta=35:00:57 +2026-08-25 05:40:00 | INFO | train rank=0 | step=4740/100000 progress=4.7% loss=0.102020 lr=9.478000e-06 elapsed=01:44:30 avg_step=1.32s recent_step=1.18s eta=35:00:15 +2026-08-25 05:40:13 | INFO | train rank=0 | step=4750/100000 progress=4.8% loss=0.091156 lr=9.498000e-06 elapsed=01:44:43 avg_step=1.32s recent_step=1.32s eta=35:00:00 +2026-08-25 05:40:26 | INFO | train rank=0 | step=4760/100000 progress=4.8% loss=0.129822 lr=9.518000e-06 elapsed=01:44:55 avg_step=1.32s recent_step=1.24s eta=34:59:31 +2026-08-25 05:40:39 | INFO | train rank=0 | step=4770/100000 progress=4.8% loss=0.103088 lr=9.538000e-06 elapsed=01:45:09 avg_step=1.32s recent_step=1.32s eta=34:59:17 +2026-08-25 05:40:53 | INFO | train rank=0 | step=4780/100000 progress=4.8% loss=0.123840 lr=9.558000e-06 elapsed=01:45:23 avg_step=1.32s recent_step=1.41s eta=34:59:20 +2026-08-25 05:41:05 | INFO | train rank=0 | step=4790/100000 progress=4.8% loss=0.065948 lr=9.578000e-06 elapsed=01:45:34 avg_step=1.32s recent_step=1.18s eta=34:58:38 +2026-08-25 05:41:20 | INFO | train rank=0 | step=4800/100000 progress=4.8% loss=0.132751 lr=9.598000e-06 elapsed=01:45:50 avg_step=1.32s recent_step=1.53s eta=34:59:06 +2026-08-25 05:41:32 | INFO | train rank=0 | step=4810/100000 progress=4.8% loss=0.104004 lr=9.618000e-06 elapsed=01:46:02 avg_step=1.32s recent_step=1.24s eta=34:58:36 +2026-08-25 05:41:45 | INFO | train rank=0 | step=4820/100000 progress=4.8% loss=0.097778 lr=9.638000e-06 elapsed=01:46:15 avg_step=1.32s recent_step=1.33s eta=34:58:23 +2026-08-25 05:41:59 | INFO | train rank=0 | step=4830/100000 progress=4.8% loss=0.079865 lr=9.658000e-06 elapsed=01:46:29 avg_step=1.32s recent_step=1.35s eta=34:58:16 +2026-08-25 05:42:11 | INFO | train rank=0 | step=4840/100000 progress=4.8% loss=0.082245 lr=9.678000e-06 elapsed=01:46:41 avg_step=1.32s recent_step=1.17s eta=34:57:33 +2026-08-25 05:42:25 | INFO | train rank=0 | step=4850/100000 progress=4.9% loss=0.106567 lr=9.698000e-06 elapsed=01:46:55 avg_step=1.32s recent_step=1.39s eta=34:57:33 +2026-08-25 05:42:38 | INFO | train rank=0 | step=4860/100000 progress=4.9% loss=0.105438 lr=9.718000e-06 elapsed=01:47:08 avg_step=1.32s recent_step=1.34s eta=34:57:23 +2026-08-25 05:42:51 | INFO | train rank=0 | step=4870/100000 progress=4.9% loss=0.085724 lr=9.738000e-06 elapsed=01:47:21 avg_step=1.32s recent_step=1.30s eta=34:57:05 +2026-08-25 05:43:11 | INFO | train rank=0 | step=4880/100000 progress=4.9% loss=0.225830 lr=9.758000e-06 elapsed=01:47:41 avg_step=1.32s recent_step=2.01s eta=34:59:05 +2026-08-25 05:43:23 | INFO | train rank=0 | step=4890/100000 progress=4.9% loss=0.109314 lr=9.778000e-06 elapsed=01:47:53 avg_step=1.32s recent_step=1.23s eta=34:58:34 +2026-08-25 05:43:35 | INFO | train rank=0 | step=4900/100000 progress=4.9% loss=0.122620 lr=9.798000e-06 elapsed=01:48:05 avg_step=1.32s recent_step=1.15s eta=34:57:46 +2026-08-25 05:43:47 | INFO | train rank=0 | step=4910/100000 progress=4.9% loss=0.118896 lr=9.818000e-06 elapsed=01:48:17 avg_step=1.32s recent_step=1.19s eta=34:57:07 +2026-08-25 05:44:07 | INFO | train rank=0 | step=4920/100000 progress=4.9% loss=0.141479 lr=9.838000e-06 elapsed=01:48:37 avg_step=1.32s recent_step=1.99s eta=34:59:03 +2026-08-25 05:44:19 | INFO | train rank=0 | step=4930/100000 progress=4.9% loss=0.062500 lr=9.858000e-06 elapsed=01:48:48 avg_step=1.32s recent_step=1.19s eta=34:58:24 +2026-08-25 05:44:31 | INFO | train rank=0 | step=4940/100000 progress=4.9% loss=0.064606 lr=9.878000e-06 elapsed=01:49:01 avg_step=1.32s recent_step=1.20s eta=34:57:48 +2026-08-25 05:44:42 | INFO | train rank=0 | step=4950/100000 progress=5.0% loss=0.110413 lr=9.898000e-06 elapsed=01:49:12 avg_step=1.32s recent_step=1.15s eta=34:57:01 +2026-08-25 05:44:59 | INFO | train rank=0 | step=4960/100000 progress=5.0% loss=0.090179 lr=9.918000e-06 elapsed=01:49:29 avg_step=1.32s recent_step=1.70s eta=34:58:00 +2026-08-25 05:45:13 | INFO | train rank=0 | step=4970/100000 progress=5.0% loss=0.094055 lr=9.938000e-06 elapsed=01:49:42 avg_step=1.32s recent_step=1.35s eta=34:57:51 +2026-08-25 05:45:24 | INFO | train rank=0 | step=4980/100000 progress=5.0% loss=0.089874 lr=9.958000e-06 elapsed=01:49:54 avg_step=1.32s recent_step=1.17s eta=34:57:08 +2026-08-25 05:45:37 | INFO | train rank=0 | step=4990/100000 progress=5.0% loss=0.071259 lr=9.978000e-06 elapsed=01:50:06 avg_step=1.32s recent_step=1.23s eta=34:56:36 +2026-08-25 05:45:49 | INFO | train rank=0 | step=5000/100000 progress=5.0% loss=0.133301 lr=9.998000e-06 elapsed=01:50:19 avg_step=1.32s recent_step=1.23s eta=34:56:06 +2026-08-25 05:46:02 | INFO | train rank=0 | step=5010/100000 progress=5.0% loss=0.072418 lr=1.000000e-05 elapsed=01:50:32 avg_step=1.32s recent_step=1.34s eta=34:55:55 +2026-08-25 05:46:14 | INFO | train rank=0 | step=5020/100000 progress=5.0% loss=0.084290 lr=9.999999e-06 elapsed=01:50:44 avg_step=1.32s recent_step=1.17s eta=34:55:13 +2026-08-25 05:46:25 | INFO | train rank=0 | step=5030/100000 progress=5.0% loss=0.147705 lr=9.999998e-06 elapsed=01:50:55 avg_step=1.32s recent_step=1.14s eta=34:54:26 +2026-08-25 05:46:38 | INFO | train rank=0 | step=5040/100000 progress=5.0% loss=0.117981 lr=9.999996e-06 elapsed=01:51:08 avg_step=1.32s recent_step=1.22s eta=34:53:53 +2026-08-25 05:46:50 | INFO | train rank=0 | step=5050/100000 progress=5.1% loss=0.084198 lr=9.999993e-06 elapsed=01:51:20 avg_step=1.32s recent_step=1.29s eta=34:53:33 +2026-08-25 05:47:02 | INFO | train rank=0 | step=5060/100000 progress=5.1% loss=0.074799 lr=9.999990e-06 elapsed=01:51:32 avg_step=1.32s recent_step=1.18s eta=34:52:53 +2026-08-25 05:47:14 | INFO | train rank=0 | step=5070/100000 progress=5.1% loss=0.073914 lr=9.999987e-06 elapsed=01:51:44 avg_step=1.32s recent_step=1.21s eta=34:52:20 +2026-08-25 05:47:26 | INFO | train rank=0 | step=5080/100000 progress=5.1% loss=0.086060 lr=9.999983e-06 elapsed=01:51:56 avg_step=1.32s recent_step=1.18s eta=34:51:41 +2026-08-25 05:47:40 | INFO | train rank=0 | step=5090/100000 progress=5.1% loss=0.153992 lr=9.999978e-06 elapsed=01:52:10 avg_step=1.32s recent_step=1.40s eta=34:51:42 +2026-08-25 05:47:55 | INFO | train rank=0 | step=5100/100000 progress=5.1% loss=0.096832 lr=9.999973e-06 elapsed=01:52:24 avg_step=1.32s recent_step=1.42s eta=34:51:48 +2026-08-25 05:48:07 | INFO | train rank=0 | step=5110/100000 progress=5.1% loss=0.104370 lr=9.999967e-06 elapsed=01:52:36 avg_step=1.32s recent_step=1.21s eta=34:51:13 +2026-08-25 05:48:26 | INFO | train rank=0 | step=5120/100000 progress=5.1% loss=0.102661 lr=9.999961e-06 elapsed=01:52:56 avg_step=1.32s recent_step=1.90s eta=34:52:48 +2026-08-25 05:48:41 | INFO | train rank=0 | step=5130/100000 progress=5.1% loss=0.102997 lr=9.999954e-06 elapsed=01:53:11 avg_step=1.32s recent_step=1.52s eta=34:53:11 +2026-08-25 05:48:53 | INFO | train rank=0 | step=5140/100000 progress=5.1% loss=0.106079 lr=9.999946e-06 elapsed=01:53:23 avg_step=1.32s recent_step=1.18s eta=34:52:31 +2026-08-25 05:49:09 | INFO | train rank=0 | step=5150/100000 progress=5.1% loss=0.069687 lr=9.999938e-06 elapsed=01:53:39 avg_step=1.32s recent_step=1.61s eta=34:53:10 +2026-08-25 05:49:20 | INFO | train rank=0 | step=5160/100000 progress=5.2% loss=0.076843 lr=9.999930e-06 elapsed=01:53:50 avg_step=1.32s recent_step=1.17s eta=34:52:28 +2026-08-25 05:49:32 | INFO | train rank=0 | step=5170/100000 progress=5.2% loss=0.083405 lr=9.999921e-06 elapsed=01:54:02 avg_step=1.32s recent_step=1.19s eta=34:51:51 +2026-08-25 05:49:45 | INFO | train rank=0 | step=5180/100000 progress=5.2% loss=0.106598 lr=9.999911e-06 elapsed=01:54:15 avg_step=1.32s recent_step=1.24s eta=34:51:23 +2026-08-25 05:50:00 | INFO | train rank=0 | step=5190/100000 progress=5.2% loss=0.093201 lr=9.999901e-06 elapsed=01:54:30 avg_step=1.32s recent_step=1.49s eta=34:51:40 +2026-08-25 05:50:13 | INFO | train rank=0 | step=5200/100000 progress=5.2% loss=0.074951 lr=9.999891e-06 elapsed=01:54:43 avg_step=1.32s recent_step=1.36s eta=34:51:33 +2026-08-25 05:50:26 | INFO | train rank=0 | step=5210/100000 progress=5.2% loss=0.070374 lr=9.999879e-06 elapsed=01:54:56 avg_step=1.32s recent_step=1.25s eta=34:51:06 +2026-08-25 05:50:37 | INFO | train rank=0 | step=5220/100000 progress=5.2% loss=0.105835 lr=9.999868e-06 elapsed=01:55:07 avg_step=1.32s recent_step=1.15s eta=34:50:21 +2026-08-25 05:50:53 | INFO | train rank=0 | step=5230/100000 progress=5.2% loss=0.071548 lr=9.999855e-06 elapsed=01:55:23 avg_step=1.32s recent_step=1.61s eta=34:51:00 +2026-08-25 05:51:08 | INFO | train rank=0 | step=5240/100000 progress=5.2% loss=0.145386 lr=9.999843e-06 elapsed=01:55:38 avg_step=1.32s recent_step=1.48s eta=34:51:15 +2026-08-25 05:51:22 | INFO | train rank=0 | step=5250/100000 progress=5.2% loss=0.151978 lr=9.999829e-06 elapsed=01:55:52 avg_step=1.32s recent_step=1.35s eta=34:51:07 +2026-08-25 05:51:34 | INFO | train rank=0 | step=5260/100000 progress=5.3% loss=0.101776 lr=9.999815e-06 elapsed=01:56:03 avg_step=1.32s recent_step=1.19s eta=34:50:30 +2026-08-25 05:51:45 | INFO | train rank=0 | step=5270/100000 progress=5.3% loss=0.073425 lr=9.999801e-06 elapsed=01:56:15 avg_step=1.32s recent_step=1.14s eta=34:49:44 +2026-08-25 05:51:57 | INFO | train rank=0 | step=5280/100000 progress=5.3% loss=0.090302 lr=9.999786e-06 elapsed=01:56:27 avg_step=1.32s recent_step=1.23s eta=34:49:14 +2026-08-25 05:52:11 | INFO | train rank=0 | step=5290/100000 progress=5.3% loss=0.113647 lr=9.999770e-06 elapsed=01:56:41 avg_step=1.32s recent_step=1.34s eta=34:49:04 +2026-08-25 05:52:22 | INFO | train rank=0 | step=5300/100000 progress=5.3% loss=0.115540 lr=9.999754e-06 elapsed=01:56:52 avg_step=1.32s recent_step=1.18s eta=34:48:24 +2026-08-25 05:52:37 | INFO | train rank=0 | step=5310/100000 progress=5.3% loss=0.087311 lr=9.999737e-06 elapsed=01:57:07 avg_step=1.32s recent_step=1.51s eta=34:48:44 +2026-08-25 05:52:50 | INFO | train rank=0 | step=5320/100000 progress=5.3% loss=0.078369 lr=9.999720e-06 elapsed=01:57:19 avg_step=1.32s recent_step=1.21s eta=34:48:10 +2026-08-25 05:53:02 | INFO | train rank=0 | step=5330/100000 progress=5.3% loss=0.112671 lr=9.999702e-06 elapsed=01:57:32 avg_step=1.32s recent_step=1.23s eta=34:47:40 +2026-08-25 05:53:15 | INFO | train rank=0 | step=5340/100000 progress=5.3% loss=0.073242 lr=9.999684e-06 elapsed=01:57:44 avg_step=1.32s recent_step=1.27s eta=34:47:17 +2026-08-25 05:53:28 | INFO | train rank=0 | step=5350/100000 progress=5.3% loss=0.076569 lr=9.999665e-06 elapsed=01:57:58 avg_step=1.32s recent_step=1.31s eta=34:47:02 +2026-08-25 05:53:40 | INFO | train rank=0 | step=5360/100000 progress=5.4% loss=0.131104 lr=9.999646e-06 elapsed=01:58:10 avg_step=1.32s recent_step=1.24s eta=34:46:34 +2026-08-25 05:53:52 | INFO | train rank=0 | step=5370/100000 progress=5.4% loss=0.101013 lr=9.999626e-06 elapsed=01:58:22 avg_step=1.32s recent_step=1.16s eta=34:45:52 +2026-08-25 05:54:04 | INFO | train rank=0 | step=5380/100000 progress=5.4% loss=0.127869 lr=9.999605e-06 elapsed=01:58:34 avg_step=1.32s recent_step=1.28s eta=34:45:31 +2026-08-25 05:54:21 | INFO | train rank=0 | step=5390/100000 progress=5.4% loss=0.067749 lr=9.999584e-06 elapsed=01:58:51 avg_step=1.32s recent_step=1.65s eta=34:46:14 +2026-08-25 05:54:36 | INFO | train rank=0 | step=5400/100000 progress=5.4% loss=0.097473 lr=9.999563e-06 elapsed=01:59:06 avg_step=1.32s recent_step=1.53s eta=34:46:37 +2026-08-25 05:54:48 | INFO | train rank=0 | step=5410/100000 progress=5.4% loss=0.077728 lr=9.999540e-06 elapsed=01:59:18 avg_step=1.32s recent_step=1.22s eta=34:46:06 +2026-08-25 05:55:02 | INFO | train rank=0 | step=5420/100000 progress=5.4% loss=0.074310 lr=9.999518e-06 elapsed=01:59:32 avg_step=1.32s recent_step=1.34s eta=34:45:56 +2026-08-25 05:55:16 | INFO | train rank=0 | step=5430/100000 progress=5.4% loss=0.090363 lr=9.999494e-06 elapsed=01:59:46 avg_step=1.32s recent_step=1.44s eta=34:46:04 +2026-08-25 05:55:29 | INFO | train rank=0 | step=5440/100000 progress=5.4% loss=0.073944 lr=9.999471e-06 elapsed=01:59:58 avg_step=1.32s recent_step=1.23s eta=34:45:34 +2026-08-25 05:55:43 | INFO | train rank=0 | step=5450/100000 progress=5.5% loss=0.083405 lr=9.999446e-06 elapsed=02:00:13 avg_step=1.32s recent_step=1.43s eta=34:45:39 +2026-08-25 05:55:58 | INFO | train rank=0 | step=5460/100000 progress=5.5% loss=0.126465 lr=9.999422e-06 elapsed=02:00:28 avg_step=1.32s recent_step=1.50s eta=34:45:57 +2026-08-25 05:56:10 | INFO | train rank=0 | step=5470/100000 progress=5.5% loss=0.068115 lr=9.999396e-06 elapsed=02:00:40 avg_step=1.32s recent_step=1.21s eta=34:45:24 +2026-08-25 05:56:26 | INFO | train rank=0 | step=5480/100000 progress=5.5% loss=0.080353 lr=9.999370e-06 elapsed=02:00:55 avg_step=1.32s recent_step=1.56s eta=34:45:52 +2026-08-25 05:56:37 | INFO | train rank=0 | step=5490/100000 progress=5.5% loss=0.059692 lr=9.999344e-06 elapsed=02:01:07 avg_step=1.32s recent_step=1.18s eta=34:45:14 +2026-08-25 05:56:55 | INFO | train rank=0 | step=5500/100000 progress=5.5% loss=0.083679 lr=9.999317e-06 elapsed=02:01:24 avg_step=1.32s recent_step=1.72s eta=34:46:08 +2026-08-25 05:57:06 | INFO | train rank=0 | step=5510/100000 progress=5.5% loss=0.060806 lr=9.999289e-06 elapsed=02:01:36 avg_step=1.32s recent_step=1.13s eta=34:45:22 +2026-08-25 05:57:18 | INFO | train rank=0 | step=5520/100000 progress=5.5% loss=0.072144 lr=9.999261e-06 elapsed=02:01:48 avg_step=1.32s recent_step=1.20s eta=34:44:48 +2026-08-25 05:57:31 | INFO | train rank=0 | step=5530/100000 progress=5.5% loss=0.126068 lr=9.999232e-06 elapsed=02:02:01 avg_step=1.32s recent_step=1.28s eta=34:44:27 +2026-08-25 05:57:42 | INFO | train rank=0 | step=5540/100000 progress=5.5% loss=0.093933 lr=9.999203e-06 elapsed=02:02:12 avg_step=1.32s recent_step=1.13s eta=34:43:41 +2026-08-25 05:57:54 | INFO | train rank=0 | step=5550/100000 progress=5.5% loss=0.080597 lr=9.999173e-06 elapsed=02:02:24 avg_step=1.32s recent_step=1.18s eta=34:43:03 +2026-08-25 05:58:09 | INFO | train rank=0 | step=5560/100000 progress=5.6% loss=0.073730 lr=9.999143e-06 elapsed=02:02:39 avg_step=1.32s recent_step=1.53s eta=34:43:25 +2026-08-25 05:58:21 | INFO | train rank=0 | step=5570/100000 progress=5.6% loss=0.105591 lr=9.999112e-06 elapsed=02:02:51 avg_step=1.32s recent_step=1.17s eta=34:42:46 +2026-08-25 05:58:33 | INFO | train rank=0 | step=5580/100000 progress=5.6% loss=0.119324 lr=9.999080e-06 elapsed=02:03:03 avg_step=1.32s recent_step=1.19s eta=34:42:11 +2026-08-25 05:58:45 | INFO | train rank=0 | step=5590/100000 progress=5.6% loss=0.104370 lr=9.999048e-06 elapsed=02:03:15 avg_step=1.32s recent_step=1.19s eta=34:41:35 +2026-08-25 05:59:04 | INFO | train rank=0 | step=5600/100000 progress=5.6% loss=0.116333 lr=9.999016e-06 elapsed=02:03:34 avg_step=1.32s recent_step=1.97s eta=34:43:12 +2026-08-25 05:59:16 | INFO | train rank=0 | step=5610/100000 progress=5.6% loss=0.141174 lr=9.998983e-06 elapsed=02:03:46 avg_step=1.32s recent_step=1.17s eta=34:42:33 +2026-08-25 05:59:28 | INFO | train rank=0 | step=5620/100000 progress=5.6% loss=0.107239 lr=9.998949e-06 elapsed=02:03:58 avg_step=1.32s recent_step=1.22s eta=34:42:02 +2026-08-25 05:59:44 | INFO | train rank=0 | step=5630/100000 progress=5.6% loss=0.063705 lr=9.998915e-06 elapsed=02:04:14 avg_step=1.32s recent_step=1.61s eta=34:42:36 +2026-08-25 05:59:56 | INFO | train rank=0 | step=5640/100000 progress=5.6% loss=0.088226 lr=9.998880e-06 elapsed=02:04:25 avg_step=1.32s recent_step=1.12s eta=34:41:48 +2026-08-25 06:00:09 | INFO | train rank=0 | step=5650/100000 progress=5.7% loss=0.064545 lr=9.998845e-06 elapsed=02:04:39 avg_step=1.32s recent_step=1.34s eta=34:41:37 +2026-08-25 06:00:22 | INFO | train rank=0 | step=5660/100000 progress=5.7% loss=0.118896 lr=9.998809e-06 elapsed=02:04:51 avg_step=1.32s recent_step=1.27s eta=34:41:15 +2026-08-25 06:00:33 | INFO | train rank=0 | step=5670/100000 progress=5.7% loss=0.160461 lr=9.998773e-06 elapsed=02:05:03 avg_step=1.32s recent_step=1.15s eta=34:40:33 +2026-08-25 06:00:46 | INFO | train rank=0 | step=5680/100000 progress=5.7% loss=0.095276 lr=9.998736e-06 elapsed=02:05:16 avg_step=1.32s recent_step=1.30s eta=34:40:17 +2026-08-25 06:01:01 | INFO | train rank=0 | step=5690/100000 progress=5.7% loss=0.090149 lr=9.998698e-06 elapsed=02:05:30 avg_step=1.32s recent_step=1.44s eta=34:40:22 +2026-08-25 06:01:15 | INFO | train rank=0 | step=5700/100000 progress=5.7% loss=0.139771 lr=9.998660e-06 elapsed=02:05:45 avg_step=1.32s recent_step=1.47s eta=34:40:33 +2026-08-25 06:01:32 | INFO | train rank=0 | step=5710/100000 progress=5.7% loss=0.081696 lr=9.998622e-06 elapsed=02:06:02 avg_step=1.32s recent_step=1.65s eta=34:41:13 +2026-08-25 06:01:43 | INFO | train rank=0 | step=5720/100000 progress=5.7% loss=0.113342 lr=9.998583e-06 elapsed=02:06:13 avg_step=1.32s recent_step=1.17s eta=34:40:33 +2026-08-25 06:01:55 | INFO | train rank=0 | step=5730/100000 progress=5.7% loss=0.084839 lr=9.998543e-06 elapsed=02:06:25 avg_step=1.32s recent_step=1.15s eta=34:39:52 +2026-08-25 06:02:07 | INFO | train rank=0 | step=5740/100000 progress=5.7% loss=0.068420 lr=9.998503e-06 elapsed=02:06:36 avg_step=1.32s recent_step=1.17s eta=34:39:13 +2026-08-25 06:02:21 | INFO | train rank=0 | step=5750/100000 progress=5.8% loss=0.093842 lr=9.998462e-06 elapsed=02:06:51 avg_step=1.32s recent_step=1.47s eta=34:39:24 +2026-08-25 06:02:36 | INFO | train rank=0 | step=5760/100000 progress=5.8% loss=0.105774 lr=9.998421e-06 elapsed=02:07:06 avg_step=1.32s recent_step=1.48s eta=34:39:36 +2026-08-25 06:02:52 | INFO | train rank=0 | step=5770/100000 progress=5.8% loss=0.064331 lr=9.998379e-06 elapsed=02:07:22 avg_step=1.32s recent_step=1.64s eta=34:40:15 +2026-08-25 06:03:06 | INFO | train rank=0 | step=5780/100000 progress=5.8% loss=0.124390 lr=9.998337e-06 elapsed=02:07:36 avg_step=1.32s recent_step=1.34s eta=34:40:04 +2026-08-25 06:03:18 | INFO | train rank=0 | step=5790/100000 progress=5.8% loss=0.066788 lr=9.998294e-06 elapsed=02:07:48 avg_step=1.32s recent_step=1.20s eta=34:39:31 +2026-08-25 06:03:30 | INFO | train rank=0 | step=5800/100000 progress=5.8% loss=0.098755 lr=9.998250e-06 elapsed=02:08:00 avg_step=1.32s recent_step=1.25s eta=34:39:05 +2026-08-25 06:03:42 | INFO | train rank=0 | step=5810/100000 progress=5.8% loss=0.082886 lr=9.998206e-06 elapsed=02:08:12 avg_step=1.32s recent_step=1.15s eta=34:38:23 +2026-08-25 06:04:02 | INFO | train rank=0 | step=5820/100000 progress=5.8% loss=0.115784 lr=9.998162e-06 elapsed=02:08:32 avg_step=1.33s recent_step=2.03s eta=34:40:05 +2026-08-25 06:04:15 | INFO | train rank=0 | step=5830/100000 progress=5.8% loss=0.076752 lr=9.998117e-06 elapsed=02:08:45 avg_step=1.33s recent_step=1.25s eta=34:39:40 +2026-08-25 06:04:27 | INFO | train rank=0 | step=5840/100000 progress=5.8% loss=0.059631 lr=9.998071e-06 elapsed=02:08:56 avg_step=1.32s recent_step=1.19s eta=34:39:05 +2026-08-25 06:04:42 | INFO | train rank=0 | step=5850/100000 progress=5.9% loss=0.124207 lr=9.998025e-06 elapsed=02:09:12 avg_step=1.33s recent_step=1.55s eta=34:39:28 +2026-08-25 06:04:54 | INFO | train rank=0 | step=5860/100000 progress=5.9% loss=0.063599 lr=9.997978e-06 elapsed=02:09:24 avg_step=1.32s recent_step=1.19s eta=34:38:52 +2026-08-25 06:05:06 | INFO | train rank=0 | step=5870/100000 progress=5.9% loss=0.142822 lr=9.997931e-06 elapsed=02:09:36 avg_step=1.32s recent_step=1.24s eta=34:38:25 +2026-08-25 06:05:18 | INFO | train rank=0 | step=5880/100000 progress=5.9% loss=0.066010 lr=9.997883e-06 elapsed=02:09:48 avg_step=1.32s recent_step=1.15s eta=34:37:44 +2026-08-25 06:05:34 | INFO | train rank=0 | step=5890/100000 progress=5.9% loss=0.127869 lr=9.997835e-06 elapsed=02:10:04 avg_step=1.32s recent_step=1.60s eta=34:38:15 +2026-08-25 06:05:46 | INFO | train rank=0 | step=5900/100000 progress=5.9% loss=0.053436 lr=9.997786e-06 elapsed=02:10:16 avg_step=1.32s recent_step=1.24s eta=34:37:48 +2026-08-25 06:06:00 | INFO | train rank=0 | step=5910/100000 progress=5.9% loss=0.088226 lr=9.997736e-06 elapsed=02:10:30 avg_step=1.33s recent_step=1.42s eta=34:37:50 +2026-08-25 06:06:16 | INFO | train rank=0 | step=5920/100000 progress=5.9% loss=0.062317 lr=9.997686e-06 elapsed=02:10:46 avg_step=1.33s recent_step=1.52s eta=34:38:07 +2026-08-25 06:06:27 | INFO | train rank=0 | step=5930/100000 progress=5.9% loss=0.117310 lr=9.997636e-06 elapsed=02:10:57 avg_step=1.33s recent_step=1.17s eta=34:37:30 +2026-08-25 06:06:39 | INFO | train rank=0 | step=5940/100000 progress=5.9% loss=0.105103 lr=9.997584e-06 elapsed=02:11:09 avg_step=1.32s recent_step=1.13s eta=34:36:45 +2026-08-25 06:06:52 | INFO | train rank=0 | step=5950/100000 progress=5.9% loss=0.085510 lr=9.997533e-06 elapsed=02:11:22 avg_step=1.32s recent_step=1.32s eta=34:36:32 +2026-08-25 06:07:03 | INFO | train rank=0 | step=5960/100000 progress=6.0% loss=0.102478 lr=9.997481e-06 elapsed=02:11:33 avg_step=1.32s recent_step=1.14s eta=34:35:49 +2026-08-25 06:07:15 | INFO | train rank=0 | step=5970/100000 progress=6.0% loss=0.111633 lr=9.997428e-06 elapsed=02:11:45 avg_step=1.32s recent_step=1.20s eta=34:35:16 +2026-08-25 06:07:28 | INFO | train rank=0 | step=5980/100000 progress=6.0% loss=0.070190 lr=9.997375e-06 elapsed=02:11:57 avg_step=1.32s recent_step=1.23s eta=34:34:49 +2026-08-25 06:07:43 | INFO | train rank=0 | step=5990/100000 progress=6.0% loss=0.071503 lr=9.997321e-06 elapsed=02:12:13 avg_step=1.32s recent_step=1.57s eta=34:35:14 +2026-08-25 06:07:55 | INFO | train rank=0 | step=6000/100000 progress=6.0% loss=0.061127 lr=9.997266e-06 elapsed=02:12:25 avg_step=1.32s recent_step=1.22s eta=34:34:45 +2026-08-25 06:08:08 | INFO | train rank=0 | step=6010/100000 progress=6.0% loss=0.121216 lr=9.997211e-06 elapsed=02:12:38 avg_step=1.32s recent_step=1.25s eta=34:34:20 +2026-08-25 06:08:20 | INFO | train rank=0 | step=6020/100000 progress=6.0% loss=0.072083 lr=9.997156e-06 elapsed=02:12:50 avg_step=1.32s recent_step=1.18s eta=34:33:45 +2026-08-25 06:08:37 | INFO | train rank=0 | step=6030/100000 progress=6.0% loss=0.109009 lr=9.997100e-06 elapsed=02:13:07 avg_step=1.32s recent_step=1.68s eta=34:34:28 +2026-08-25 06:08:48 | INFO | train rank=0 | step=6040/100000 progress=6.0% loss=0.061615 lr=9.997043e-06 elapsed=02:13:18 avg_step=1.32s recent_step=1.13s eta=34:33:44 +2026-08-25 06:09:00 | INFO | train rank=0 | step=6050/100000 progress=6.0% loss=0.111450 lr=9.996986e-06 elapsed=02:13:30 avg_step=1.32s recent_step=1.21s eta=34:33:13 +2026-08-25 06:09:13 | INFO | train rank=0 | step=6060/100000 progress=6.1% loss=0.126892 lr=9.996928e-06 elapsed=02:13:43 avg_step=1.32s recent_step=1.30s eta=34:32:57 +2026-08-25 06:09:25 | INFO | train rank=0 | step=6070/100000 progress=6.1% loss=0.075897 lr=9.996870e-06 elapsed=02:13:55 avg_step=1.32s recent_step=1.18s eta=34:32:21 +2026-08-25 06:09:43 | INFO | train rank=0 | step=6080/100000 progress=6.1% loss=0.079315 lr=9.996811e-06 elapsed=02:14:13 avg_step=1.32s recent_step=1.85s eta=34:33:29 +2026-08-25 06:09:55 | INFO | train rank=0 | step=6090/100000 progress=6.1% loss=0.108887 lr=9.996752e-06 elapsed=02:14:25 avg_step=1.32s recent_step=1.13s eta=34:32:46 +2026-08-25 06:10:09 | INFO | train rank=0 | step=6100/100000 progress=6.1% loss=0.128784 lr=9.996692e-06 elapsed=02:14:39 avg_step=1.32s recent_step=1.40s eta=34:32:44 +2026-08-25 06:10:21 | INFO | train rank=0 | step=6110/100000 progress=6.1% loss=0.077423 lr=9.996632e-06 elapsed=02:14:51 avg_step=1.32s recent_step=1.22s eta=34:32:15 +2026-08-25 06:10:33 | INFO | train rank=0 | step=6120/100000 progress=6.1% loss=0.095734 lr=9.996571e-06 elapsed=02:15:03 avg_step=1.32s recent_step=1.21s eta=34:31:44 +2026-08-25 06:10:46 | INFO | train rank=0 | step=6130/100000 progress=6.1% loss=0.099182 lr=9.996509e-06 elapsed=02:15:15 avg_step=1.32s recent_step=1.25s eta=34:31:20 +2026-08-25 06:10:57 | INFO | train rank=0 | step=6140/100000 progress=6.1% loss=0.076920 lr=9.996447e-06 elapsed=02:15:27 avg_step=1.32s recent_step=1.18s eta=34:30:45 +2026-08-25 06:11:09 | INFO | train rank=0 | step=6150/100000 progress=6.2% loss=0.082001 lr=9.996385e-06 elapsed=02:15:39 avg_step=1.32s recent_step=1.17s eta=34:30:09 +2026-08-25 06:11:23 | INFO | train rank=0 | step=6160/100000 progress=6.2% loss=0.071991 lr=9.996322e-06 elapsed=02:15:53 avg_step=1.32s recent_step=1.40s eta=34:30:07 +2026-08-25 06:11:37 | INFO | train rank=0 | step=6170/100000 progress=6.2% loss=0.084198 lr=9.996258e-06 elapsed=02:16:07 avg_step=1.32s recent_step=1.36s eta=34:30:00 +2026-08-25 06:11:49 | INFO | train rank=0 | step=6180/100000 progress=6.2% loss=0.070557 lr=9.996194e-06 elapsed=02:16:19 avg_step=1.32s recent_step=1.24s eta=34:29:35 +2026-08-25 06:12:01 | INFO | train rank=0 | step=6190/100000 progress=6.2% loss=0.140198 lr=9.996129e-06 elapsed=02:16:31 avg_step=1.32s recent_step=1.17s eta=34:28:58 +2026-08-25 06:12:13 | INFO | train rank=0 | step=6200/100000 progress=6.2% loss=0.077759 lr=9.996064e-06 elapsed=02:16:43 avg_step=1.32s recent_step=1.25s eta=34:28:35 +2026-08-25 06:12:25 | INFO | train rank=0 | step=6210/100000 progress=6.2% loss=0.071121 lr=9.995998e-06 elapsed=02:16:55 avg_step=1.32s recent_step=1.20s eta=34:28:02 +2026-08-25 06:12:38 | INFO | train rank=0 | step=6220/100000 progress=6.2% loss=0.078430 lr=9.995931e-06 elapsed=02:17:08 avg_step=1.32s recent_step=1.30s eta=34:27:46 +2026-08-25 06:12:50 | INFO | train rank=0 | step=6230/100000 progress=6.2% loss=0.119995 lr=9.995864e-06 elapsed=02:17:20 avg_step=1.32s recent_step=1.16s eta=34:27:09 +2026-08-25 06:13:06 | INFO | train rank=0 | step=6240/100000 progress=6.2% loss=0.115601 lr=9.995797e-06 elapsed=02:17:36 avg_step=1.32s recent_step=1.59s eta=34:27:36 +2026-08-25 06:13:20 | INFO | train rank=0 | step=6250/100000 progress=6.2% loss=0.104370 lr=9.995729e-06 elapsed=02:17:50 avg_step=1.32s recent_step=1.40s eta=34:27:35 +2026-08-25 06:13:34 | INFO | train rank=0 | step=6260/100000 progress=6.3% loss=0.082886 lr=9.995660e-06 elapsed=02:18:04 avg_step=1.32s recent_step=1.45s eta=34:27:40 +2026-08-25 06:13:46 | INFO | train rank=0 | step=6270/100000 progress=6.3% loss=0.116821 lr=9.995591e-06 elapsed=02:18:16 avg_step=1.32s recent_step=1.18s eta=34:27:06 +2026-08-25 06:13:58 | INFO | train rank=0 | step=6280/100000 progress=6.3% loss=0.076263 lr=9.995521e-06 elapsed=02:18:28 avg_step=1.32s recent_step=1.19s eta=34:26:33 +2026-08-25 06:14:10 | INFO | train rank=0 | step=6290/100000 progress=6.3% loss=0.081268 lr=9.995451e-06 elapsed=02:18:40 avg_step=1.32s recent_step=1.21s eta=34:26:02 +2026-08-25 06:14:23 | INFO | train rank=0 | step=6300/100000 progress=6.3% loss=0.077057 lr=9.995380e-06 elapsed=02:18:53 avg_step=1.32s recent_step=1.30s eta=34:25:46 +2026-08-25 06:14:35 | INFO | train rank=0 | step=6310/100000 progress=6.3% loss=0.091797 lr=9.995309e-06 elapsed=02:19:05 avg_step=1.32s recent_step=1.17s eta=34:25:10 +2026-08-25 06:14:51 | INFO | train rank=0 | step=6320/100000 progress=6.3% loss=0.137573 lr=9.995237e-06 elapsed=02:19:21 avg_step=1.32s recent_step=1.59s eta=34:25:37 +2026-08-25 06:15:02 | INFO | train rank=0 | step=6330/100000 progress=6.3% loss=0.094147 lr=9.995165e-06 elapsed=02:19:32 avg_step=1.32s recent_step=1.14s eta=34:24:57 +2026-08-25 06:15:15 | INFO | train rank=0 | step=6340/100000 progress=6.3% loss=0.089966 lr=9.995092e-06 elapsed=02:19:45 avg_step=1.32s recent_step=1.26s eta=34:24:34 +2026-08-25 06:15:28 | INFO | train rank=0 | step=6350/100000 progress=6.3% loss=0.121033 lr=9.995018e-06 elapsed=02:19:58 avg_step=1.32s recent_step=1.28s eta=34:24:14 +2026-08-25 06:15:41 | INFO | train rank=0 | step=6360/100000 progress=6.4% loss=0.070496 lr=9.994944e-06 elapsed=02:20:10 avg_step=1.32s recent_step=1.28s eta=34:23:56 +2026-08-25 06:15:55 | INFO | train rank=0 | step=6370/100000 progress=6.4% loss=0.076263 lr=9.994870e-06 elapsed=02:20:25 avg_step=1.32s recent_step=1.46s eta=34:24:03 +2026-08-25 06:16:08 | INFO | train rank=0 | step=6380/100000 progress=6.4% loss=0.076385 lr=9.994794e-06 elapsed=02:20:37 avg_step=1.32s recent_step=1.24s eta=34:23:38 +2026-08-25 06:16:26 | INFO | train rank=0 | step=6390/100000 progress=6.4% loss=0.085083 lr=9.994719e-06 elapsed=02:20:56 avg_step=1.32s recent_step=1.87s eta=34:24:45 +2026-08-25 06:16:38 | INFO | train rank=0 | step=6400/100000 progress=6.4% loss=0.100861 lr=9.994642e-06 elapsed=02:21:08 avg_step=1.32s recent_step=1.19s eta=34:24:12 +2026-08-25 06:16:54 | INFO | train rank=0 | step=6410/100000 progress=6.4% loss=0.104919 lr=9.994566e-06 elapsed=02:21:24 avg_step=1.32s recent_step=1.58s eta=34:24:37 +2026-08-25 06:17:06 | INFO | train rank=0 | step=6420/100000 progress=6.4% loss=0.111511 lr=9.994488e-06 elapsed=02:21:36 avg_step=1.32s recent_step=1.25s eta=34:24:13 +2026-08-25 06:17:19 | INFO | train rank=0 | step=6430/100000 progress=6.4% loss=0.076904 lr=9.994410e-06 elapsed=02:21:49 avg_step=1.32s recent_step=1.27s eta=34:23:52 +2026-08-25 06:17:37 | INFO | train rank=0 | step=6440/100000 progress=6.4% loss=0.076782 lr=9.994332e-06 elapsed=02:22:07 avg_step=1.32s recent_step=1.83s eta=34:24:52 +2026-08-25 06:17:49 | INFO | train rank=0 | step=6450/100000 progress=6.5% loss=0.098816 lr=9.994253e-06 elapsed=02:22:19 avg_step=1.32s recent_step=1.18s eta=34:24:18 +2026-08-25 06:18:05 | INFO | train rank=0 | step=6460/100000 progress=6.5% loss=0.062073 lr=9.994173e-06 elapsed=02:22:35 avg_step=1.32s recent_step=1.54s eta=34:24:35 +2026-08-25 06:18:18 | INFO | train rank=0 | step=6470/100000 progress=6.5% loss=0.112000 lr=9.994093e-06 elapsed=02:22:48 avg_step=1.32s recent_step=1.37s eta=34:24:28 +2026-08-25 06:18:30 | INFO | train rank=0 | step=6480/100000 progress=6.5% loss=0.062103 lr=9.994013e-06 elapsed=02:23:00 avg_step=1.32s recent_step=1.16s eta=34:23:52 +2026-08-25 06:18:41 | INFO | train rank=0 | step=6490/100000 progress=6.5% loss=0.060181 lr=9.993932e-06 elapsed=02:23:11 avg_step=1.32s recent_step=1.15s eta=34:23:14 +2026-08-25 06:18:54 | INFO | train rank=0 | step=6500/100000 progress=6.5% loss=0.122681 lr=9.993850e-06 elapsed=02:23:24 avg_step=1.32s recent_step=1.21s eta=34:22:45 +2026-08-25 06:19:06 | INFO | train rank=0 | step=6510/100000 progress=6.5% loss=0.066528 lr=9.993768e-06 elapsed=02:23:36 avg_step=1.32s recent_step=1.26s eta=34:22:22 +2026-08-25 06:19:19 | INFO | train rank=0 | step=6520/100000 progress=6.5% loss=0.179810 lr=9.993685e-06 elapsed=02:23:49 avg_step=1.32s recent_step=1.26s eta=34:22:00 +2026-08-25 06:19:36 | INFO | train rank=0 | step=6530/100000 progress=6.5% loss=0.096558 lr=9.993601e-06 elapsed=02:24:06 avg_step=1.32s recent_step=1.71s eta=34:22:42 +2026-08-25 06:19:48 | INFO | train rank=0 | step=6540/100000 progress=6.5% loss=0.062469 lr=9.993518e-06 elapsed=02:24:18 avg_step=1.32s recent_step=1.17s eta=34:22:07 +2026-08-25 06:20:02 | INFO | train rank=0 | step=6550/100000 progress=6.6% loss=0.062073 lr=9.993433e-06 elapsed=02:24:32 avg_step=1.32s recent_step=1.43s eta=34:22:08 +2026-08-25 06:20:16 | INFO | train rank=0 | step=6560/100000 progress=6.6% loss=0.084900 lr=9.993348e-06 elapsed=02:24:45 avg_step=1.32s recent_step=1.37s eta=34:22:01 +2026-08-25 06:20:28 | INFO | train rank=0 | step=6570/100000 progress=6.6% loss=0.085327 lr=9.993263e-06 elapsed=02:24:58 avg_step=1.32s recent_step=1.24s eta=34:21:36 +2026-08-25 06:20:40 | INFO | train rank=0 | step=6580/100000 progress=6.6% loss=0.093933 lr=9.993176e-06 elapsed=02:25:10 avg_step=1.32s recent_step=1.23s eta=34:21:10 +2026-08-25 06:20:54 | INFO | train rank=0 | step=6590/100000 progress=6.6% loss=0.092010 lr=9.993090e-06 elapsed=02:25:24 avg_step=1.32s recent_step=1.34s eta=34:20:59 +2026-08-25 06:21:09 | INFO | train rank=0 | step=6600/100000 progress=6.6% loss=0.124084 lr=9.993003e-06 elapsed=02:25:39 avg_step=1.32s recent_step=1.54s eta=34:21:17 +2026-08-25 06:21:21 | INFO | train rank=0 | step=6610/100000 progress=6.6% loss=0.090393 lr=9.992915e-06 elapsed=02:25:51 avg_step=1.32s recent_step=1.21s eta=34:20:47 +2026-08-25 06:21:35 | INFO | train rank=0 | step=6620/100000 progress=6.6% loss=0.064240 lr=9.992827e-06 elapsed=02:26:05 avg_step=1.32s recent_step=1.37s eta=34:20:40 +2026-08-25 06:21:47 | INFO | train rank=0 | step=6630/100000 progress=6.6% loss=0.063919 lr=9.992738e-06 elapsed=02:26:17 avg_step=1.32s recent_step=1.22s eta=34:20:13 +2026-08-25 06:22:03 | INFO | train rank=0 | step=6640/100000 progress=6.6% loss=0.124573 lr=9.992649e-06 elapsed=02:26:33 avg_step=1.32s recent_step=1.59s eta=34:20:37 +2026-08-25 06:22:15 | INFO | train rank=0 | step=6650/100000 progress=6.7% loss=0.066986 lr=9.992559e-06 elapsed=02:26:45 avg_step=1.32s recent_step=1.18s eta=34:20:03 +2026-08-25 06:22:27 | INFO | train rank=0 | step=6660/100000 progress=6.7% loss=0.058441 lr=9.992468e-06 elapsed=02:26:57 avg_step=1.32s recent_step=1.23s eta=34:19:37 +2026-08-25 06:22:39 | INFO | train rank=0 | step=6670/100000 progress=6.7% loss=0.075531 lr=9.992377e-06 elapsed=02:27:09 avg_step=1.32s recent_step=1.20s eta=34:19:06 +2026-08-25 06:22:55 | INFO | train rank=0 | step=6680/100000 progress=6.7% loss=0.116882 lr=9.992286e-06 elapsed=02:27:25 avg_step=1.32s recent_step=1.62s eta=34:19:34 +2026-08-25 06:23:07 | INFO | train rank=0 | step=6690/100000 progress=6.7% loss=0.082611 lr=9.992194e-06 elapsed=02:27:37 avg_step=1.32s recent_step=1.15s eta=34:18:58 +2026-08-25 06:23:19 | INFO | train rank=0 | step=6700/100000 progress=6.7% loss=0.067932 lr=9.992101e-06 elapsed=02:27:49 avg_step=1.32s recent_step=1.26s eta=34:18:36 +2026-08-25 06:23:36 | INFO | train rank=0 | step=6710/100000 progress=6.7% loss=0.067413 lr=9.992008e-06 elapsed=02:28:05 avg_step=1.32s recent_step=1.61s eta=34:19:03 +2026-08-25 06:23:50 | INFO | train rank=0 | step=6720/100000 progress=6.7% loss=0.131714 lr=9.991914e-06 elapsed=02:28:20 avg_step=1.32s recent_step=1.47s eta=34:19:10 +2026-08-25 06:24:02 | INFO | train rank=0 | step=6730/100000 progress=6.7% loss=0.092377 lr=9.991820e-06 elapsed=02:28:32 avg_step=1.32s recent_step=1.16s eta=34:18:34 +2026-08-25 06:24:16 | INFO | train rank=0 | step=6740/100000 progress=6.7% loss=0.084045 lr=9.991725e-06 elapsed=02:28:46 avg_step=1.32s recent_step=1.42s eta=34:18:34 +2026-08-25 06:24:28 | INFO | train rank=0 | step=6750/100000 progress=6.8% loss=0.056885 lr=9.991630e-06 elapsed=02:28:58 avg_step=1.32s recent_step=1.18s eta=34:18:01 +2026-08-25 06:24:42 | INFO | train rank=0 | step=6760/100000 progress=6.8% loss=0.083496 lr=9.991534e-06 elapsed=02:29:11 avg_step=1.32s recent_step=1.36s eta=34:17:53 +2026-08-25 06:24:54 | INFO | train rank=0 | step=6770/100000 progress=6.8% loss=0.080963 lr=9.991437e-06 elapsed=02:29:24 avg_step=1.32s recent_step=1.21s eta=34:17:25 +2026-08-25 06:25:06 | INFO | train rank=0 | step=6780/100000 progress=6.8% loss=0.051193 lr=9.991340e-06 elapsed=02:29:35 avg_step=1.32s recent_step=1.18s eta=34:16:52 +2026-08-25 06:25:21 | INFO | train rank=0 | step=6790/100000 progress=6.8% loss=0.073410 lr=9.991243e-06 elapsed=02:29:51 avg_step=1.32s recent_step=1.57s eta=34:17:13 +2026-08-25 06:25:33 | INFO | train rank=0 | step=6800/100000 progress=6.8% loss=0.083954 lr=9.991145e-06 elapsed=02:30:03 avg_step=1.32s recent_step=1.19s eta=34:16:41 +2026-08-25 06:25:46 | INFO | train rank=0 | step=6810/100000 progress=6.8% loss=0.087433 lr=9.991046e-06 elapsed=02:30:15 avg_step=1.32s recent_step=1.25s eta=34:16:17 +2026-08-25 06:25:59 | INFO | train rank=0 | step=6820/100000 progress=6.8% loss=0.064697 lr=9.990947e-06 elapsed=02:30:29 avg_step=1.32s recent_step=1.33s eta=34:16:05 +2026-08-25 06:26:13 | INFO | train rank=0 | step=6830/100000 progress=6.8% loss=0.045242 lr=9.990847e-06 elapsed=02:30:43 avg_step=1.32s recent_step=1.42s eta=34:16:05 +2026-08-25 06:26:26 | INFO | train rank=0 | step=6840/100000 progress=6.8% loss=0.069672 lr=9.990747e-06 elapsed=02:30:56 avg_step=1.32s recent_step=1.30s eta=34:15:49 +2026-08-25 06:26:40 | INFO | train rank=0 | step=6850/100000 progress=6.9% loss=0.080780 lr=9.990646e-06 elapsed=02:31:10 avg_step=1.32s recent_step=1.38s eta=34:15:43 +2026-08-25 06:26:56 | INFO | train rank=0 | step=6860/100000 progress=6.9% loss=0.069855 lr=9.990545e-06 elapsed=02:31:26 avg_step=1.32s recent_step=1.61s eta=34:16:09 +2026-08-25 06:27:10 | INFO | train rank=0 | step=6870/100000 progress=6.9% loss=0.083405 lr=9.990443e-06 elapsed=02:31:40 avg_step=1.32s recent_step=1.38s eta=34:16:03 +2026-08-25 06:27:22 | INFO | train rank=0 | step=6880/100000 progress=6.9% loss=0.092987 lr=9.990340e-06 elapsed=02:31:52 avg_step=1.32s recent_step=1.19s eta=34:15:31 +2026-08-25 06:27:34 | INFO | train rank=0 | step=6890/100000 progress=6.9% loss=0.099854 lr=9.990237e-06 elapsed=02:32:04 avg_step=1.32s recent_step=1.23s eta=34:15:05 +2026-08-25 06:27:46 | INFO | train rank=0 | step=6900/100000 progress=6.9% loss=0.068634 lr=9.990134e-06 elapsed=02:32:16 avg_step=1.32s recent_step=1.22s eta=34:14:37 +2026-08-25 06:27:58 | INFO | train rank=0 | step=6910/100000 progress=6.9% loss=0.052673 lr=9.990030e-06 elapsed=02:32:28 avg_step=1.32s recent_step=1.20s eta=34:14:07 +2026-08-25 06:28:10 | INFO | train rank=0 | step=6920/100000 progress=6.9% loss=0.143433 lr=9.989925e-06 elapsed=02:32:40 avg_step=1.32s recent_step=1.21s eta=34:13:38 +2026-08-25 06:28:22 | INFO | train rank=0 | step=6930/100000 progress=6.9% loss=0.115662 lr=9.989820e-06 elapsed=02:32:52 avg_step=1.32s recent_step=1.22s eta=34:13:11 +2026-08-25 06:28:37 | INFO | train rank=0 | step=6940/100000 progress=6.9% loss=0.077271 lr=9.989714e-06 elapsed=02:33:07 avg_step=1.32s recent_step=1.44s eta=34:13:14 +2026-08-25 06:28:49 | INFO | train rank=0 | step=6950/100000 progress=7.0% loss=0.079041 lr=9.989608e-06 elapsed=02:33:19 avg_step=1.32s recent_step=1.21s eta=34:12:45 +2026-08-25 06:29:02 | INFO | train rank=0 | step=6960/100000 progress=7.0% loss=0.067505 lr=9.989501e-06 elapsed=02:33:31 avg_step=1.32s recent_step=1.26s eta=34:12:22 +2026-08-25 06:29:14 | INFO | train rank=0 | step=6970/100000 progress=7.0% loss=0.064758 lr=9.989394e-06 elapsed=02:33:44 avg_step=1.32s recent_step=1.28s eta=34:12:03 +2026-08-25 06:29:29 | INFO | train rank=0 | step=6980/100000 progress=7.0% loss=0.091034 lr=9.989286e-06 elapsed=02:33:59 avg_step=1.32s recent_step=1.50s eta=34:12:13 +2026-08-25 06:29:41 | INFO | train rank=0 | step=6990/100000 progress=7.0% loss=0.073792 lr=9.989177e-06 elapsed=02:34:11 avg_step=1.32s recent_step=1.19s eta=34:11:43 +2026-08-25 06:29:56 | INFO | train rank=0 | step=7000/100000 progress=7.0% loss=0.095642 lr=9.989068e-06 elapsed=02:34:26 avg_step=1.32s recent_step=1.50s eta=34:11:52 +2026-08-25 06:30:08 | INFO | train rank=0 | step=7010/100000 progress=7.0% loss=0.064362 lr=9.988959e-06 elapsed=02:34:38 avg_step=1.32s recent_step=1.18s eta=34:11:20 +2026-08-25 06:30:20 | INFO | train rank=0 | step=7020/100000 progress=7.0% loss=0.110779 lr=9.988848e-06 elapsed=02:34:50 avg_step=1.32s recent_step=1.23s eta=34:10:54 +2026-08-25 06:30:35 | INFO | train rank=0 | step=7030/100000 progress=7.0% loss=0.061737 lr=9.988738e-06 elapsed=02:35:05 avg_step=1.32s recent_step=1.44s eta=34:10:57 +2026-08-25 06:30:48 | INFO | train rank=0 | step=7040/100000 progress=7.0% loss=0.114929 lr=9.988627e-06 elapsed=02:35:18 avg_step=1.32s recent_step=1.32s eta=34:10:43 +2026-08-25 06:31:02 | INFO | train rank=0 | step=7050/100000 progress=7.0% loss=0.093231 lr=9.988515e-06 elapsed=02:35:32 avg_step=1.32s recent_step=1.41s eta=34:10:42 +2026-08-25 06:31:14 | INFO | train rank=0 | step=7060/100000 progress=7.1% loss=0.146179 lr=9.988403e-06 elapsed=02:35:44 avg_step=1.32s recent_step=1.21s eta=34:10:14 +2026-08-25 06:31:29 | INFO | train rank=0 | step=7070/100000 progress=7.1% loss=0.069611 lr=9.988290e-06 elapsed=02:35:59 avg_step=1.32s recent_step=1.52s eta=34:10:27 +2026-08-25 06:31:42 | INFO | train rank=0 | step=7080/100000 progress=7.1% loss=0.071442 lr=9.988176e-06 elapsed=02:36:12 avg_step=1.32s recent_step=1.22s eta=34:10:01 +2026-08-25 06:31:54 | INFO | train rank=0 | step=7090/100000 progress=7.1% loss=0.134521 lr=9.988063e-06 elapsed=02:36:24 avg_step=1.32s recent_step=1.26s eta=34:09:39 +2026-08-25 06:32:09 | INFO | train rank=0 | step=7100/100000 progress=7.1% loss=0.108398 lr=9.987948e-06 elapsed=02:36:39 avg_step=1.32s recent_step=1.46s eta=34:09:44 +2026-08-25 06:32:25 | INFO | train rank=0 | step=7110/100000 progress=7.1% loss=0.063507 lr=9.987833e-06 elapsed=02:36:54 avg_step=1.32s recent_step=1.57s eta=34:10:02 +2026-08-25 06:32:37 | INFO | train rank=0 | step=7120/100000 progress=7.1% loss=0.075287 lr=9.987718e-06 elapsed=02:37:06 avg_step=1.32s recent_step=1.21s eta=34:09:34 +2026-08-25 06:32:49 | INFO | train rank=0 | step=7130/100000 progress=7.1% loss=0.083832 lr=9.987601e-06 elapsed=02:37:19 avg_step=1.32s recent_step=1.24s eta=34:09:10 +2026-08-25 06:33:04 | INFO | train rank=0 | step=7140/100000 progress=7.1% loss=0.053787 lr=9.987485e-06 elapsed=02:37:34 avg_step=1.32s recent_step=1.53s eta=34:09:24 +2026-08-25 06:33:16 | INFO | train rank=0 | step=7150/100000 progress=7.1% loss=0.082642 lr=9.987368e-06 elapsed=02:37:46 avg_step=1.32s recent_step=1.21s eta=34:08:55 +2026-08-25 06:33:31 | INFO | train rank=0 | step=7160/100000 progress=7.2% loss=0.078156 lr=9.987250e-06 elapsed=02:38:01 avg_step=1.32s recent_step=1.50s eta=34:09:05 +2026-08-25 06:33:43 | INFO | train rank=0 | step=7170/100000 progress=7.2% loss=0.069763 lr=9.987132e-06 elapsed=02:38:13 avg_step=1.32s recent_step=1.19s eta=34:08:35 +2026-08-25 06:33:59 | INFO | train rank=0 | step=7180/100000 progress=7.2% loss=0.103699 lr=9.987013e-06 elapsed=02:38:29 avg_step=1.32s recent_step=1.56s eta=34:08:53 +2026-08-25 06:34:11 | INFO | train rank=0 | step=7190/100000 progress=7.2% loss=0.077911 lr=9.986893e-06 elapsed=02:38:41 avg_step=1.32s recent_step=1.23s eta=34:08:28 +2026-08-25 06:34:26 | INFO | train rank=0 | step=7200/100000 progress=7.2% loss=0.094482 lr=9.986773e-06 elapsed=02:38:56 avg_step=1.32s recent_step=1.44s eta=34:08:30 +2026-08-25 06:34:37 | INFO | train rank=0 | step=7210/100000 progress=7.2% loss=0.077667 lr=9.986653e-06 elapsed=02:39:07 avg_step=1.32s recent_step=1.15s eta=34:07:55 +2026-08-25 06:34:52 | INFO | train rank=0 | step=7220/100000 progress=7.2% loss=0.061554 lr=9.986532e-06 elapsed=02:39:22 avg_step=1.32s recent_step=1.44s eta=34:07:56 +2026-08-25 06:35:06 | INFO | train rank=0 | step=7230/100000 progress=7.2% loss=0.097778 lr=9.986410e-06 elapsed=02:39:36 avg_step=1.32s recent_step=1.40s eta=34:07:52 +2026-08-25 06:35:18 | INFO | train rank=0 | step=7240/100000 progress=7.2% loss=0.125366 lr=9.986288e-06 elapsed=02:39:48 avg_step=1.32s recent_step=1.23s eta=34:07:28 +2026-08-25 06:35:34 | INFO | train rank=0 | step=7250/100000 progress=7.2% loss=0.173950 lr=9.986166e-06 elapsed=02:40:04 avg_step=1.32s recent_step=1.61s eta=34:07:50 +2026-08-25 06:35:47 | INFO | train rank=0 | step=7260/100000 progress=7.3% loss=0.096039 lr=9.986043e-06 elapsed=02:40:17 avg_step=1.32s recent_step=1.29s eta=34:07:33 +2026-08-25 06:36:03 | INFO | train rank=0 | step=7270/100000 progress=7.3% loss=0.063751 lr=9.985919e-06 elapsed=02:40:33 avg_step=1.33s recent_step=1.60s eta=34:07:55 +2026-08-25 06:36:15 | INFO | train rank=0 | step=7280/100000 progress=7.3% loss=0.127258 lr=9.985795e-06 elapsed=02:40:45 avg_step=1.32s recent_step=1.16s eta=34:07:21 +2026-08-25 06:36:26 | INFO | train rank=0 | step=7290/100000 progress=7.3% loss=0.111267 lr=9.985670e-06 elapsed=02:40:56 avg_step=1.32s recent_step=1.15s eta=34:06:46 +2026-08-25 06:36:39 | INFO | train rank=0 | step=7300/100000 progress=7.3% loss=0.086151 lr=9.985544e-06 elapsed=02:41:09 avg_step=1.32s recent_step=1.31s eta=34:06:30 +2026-08-25 06:36:53 | INFO | train rank=0 | step=7310/100000 progress=7.3% loss=0.069244 lr=9.985418e-06 elapsed=02:41:23 avg_step=1.32s recent_step=1.38s eta=34:06:24 +2026-08-25 06:37:08 | INFO | train rank=0 | step=7320/100000 progress=7.3% loss=0.075531 lr=9.985292e-06 elapsed=02:41:37 avg_step=1.32s recent_step=1.46s eta=34:06:28 +2026-08-25 06:37:21 | INFO | train rank=0 | step=7330/100000 progress=7.3% loss=0.070709 lr=9.985165e-06 elapsed=02:41:51 avg_step=1.32s recent_step=1.31s eta=34:06:12 +2026-08-25 06:37:34 | INFO | train rank=0 | step=7340/100000 progress=7.3% loss=0.089844 lr=9.985037e-06 elapsed=02:42:04 avg_step=1.32s recent_step=1.34s eta=34:06:01 +2026-08-25 06:37:46 | INFO | train rank=0 | step=7350/100000 progress=7.3% loss=0.062378 lr=9.984909e-06 elapsed=02:42:15 avg_step=1.32s recent_step=1.15s eta=34:05:25 +2026-08-25 06:37:57 | INFO | train rank=0 | step=7360/100000 progress=7.4% loss=0.078613 lr=9.984781e-06 elapsed=02:42:27 avg_step=1.32s recent_step=1.18s eta=34:04:54 +2026-08-25 06:38:10 | INFO | train rank=0 | step=7370/100000 progress=7.4% loss=0.139648 lr=9.984651e-06 elapsed=02:42:40 avg_step=1.32s recent_step=1.23s eta=34:04:29 +2026-08-25 06:38:25 | INFO | train rank=0 | step=7380/100000 progress=7.4% loss=0.061859 lr=9.984522e-06 elapsed=02:42:55 avg_step=1.32s recent_step=1.55s eta=34:04:44 +2026-08-25 06:38:37 | INFO | train rank=0 | step=7390/100000 progress=7.4% loss=0.078278 lr=9.984391e-06 elapsed=02:43:07 avg_step=1.32s recent_step=1.19s eta=34:04:13 +2026-08-25 06:38:49 | INFO | train rank=0 | step=7400/100000 progress=7.4% loss=0.119873 lr=9.984261e-06 elapsed=02:43:19 avg_step=1.32s recent_step=1.17s eta=34:03:41 +2026-08-25 06:39:01 | INFO | train rank=0 | step=7410/100000 progress=7.4% loss=0.138733 lr=9.984129e-06 elapsed=02:43:30 avg_step=1.32s recent_step=1.18s eta=34:03:10 +2026-08-25 06:39:14 | INFO | train rank=0 | step=7420/100000 progress=7.4% loss=0.081116 lr=9.983997e-06 elapsed=02:43:43 avg_step=1.32s recent_step=1.30s eta=34:02:53 +2026-08-25 06:39:26 | INFO | train rank=0 | step=7430/100000 progress=7.4% loss=0.140442 lr=9.983865e-06 elapsed=02:43:56 avg_step=1.32s recent_step=1.25s eta=34:02:31 +2026-08-25 06:39:37 | INFO | train rank=0 | step=7440/100000 progress=7.4% loss=0.067612 lr=9.983732e-06 elapsed=02:44:07 avg_step=1.32s recent_step=1.15s eta=34:01:56 +2026-08-25 06:39:50 | INFO | train rank=0 | step=7450/100000 progress=7.4% loss=0.066681 lr=9.983598e-06 elapsed=02:44:19 avg_step=1.32s recent_step=1.21s eta=34:01:28 +2026-08-25 06:40:04 | INFO | train rank=0 | step=7460/100000 progress=7.5% loss=0.119690 lr=9.983464e-06 elapsed=02:44:34 avg_step=1.32s recent_step=1.42s eta=34:01:27 +2026-08-25 06:40:15 | INFO | train rank=0 | step=7470/100000 progress=7.5% loss=0.071564 lr=9.983330e-06 elapsed=02:44:45 avg_step=1.32s recent_step=1.16s eta=34:00:54 +2026-08-25 06:40:31 | INFO | train rank=0 | step=7480/100000 progress=7.5% loss=0.083832 lr=9.983194e-06 elapsed=02:45:00 avg_step=1.32s recent_step=1.52s eta=34:01:05 +2026-08-25 06:40:43 | INFO | train rank=0 | step=7490/100000 progress=7.5% loss=0.080688 lr=9.983059e-06 elapsed=02:45:13 avg_step=1.32s recent_step=1.26s eta=34:00:44 +2026-08-25 06:40:55 | INFO | train rank=0 | step=7500/100000 progress=7.5% loss=0.058502 lr=9.982922e-06 elapsed=02:45:25 avg_step=1.32s recent_step=1.18s eta=34:00:13 +2026-08-25 06:41:09 | INFO | train rank=0 | step=7510/100000 progress=7.5% loss=0.124084 lr=9.982786e-06 elapsed=02:45:38 avg_step=1.32s recent_step=1.35s eta=34:00:04 +2026-08-25 06:41:21 | INFO | train rank=0 | step=7520/100000 progress=7.5% loss=0.073883 lr=9.982648e-06 elapsed=02:45:50 avg_step=1.32s recent_step=1.20s eta=33:59:35 +2026-08-25 06:41:37 | INFO | train rank=0 | step=7530/100000 progress=7.5% loss=0.122620 lr=9.982510e-06 elapsed=02:46:07 avg_step=1.32s recent_step=1.67s eta=34:00:05 +2026-08-25 06:41:51 | INFO | train rank=0 | step=7540/100000 progress=7.5% loss=0.106628 lr=9.982372e-06 elapsed=02:46:21 avg_step=1.32s recent_step=1.35s eta=33:59:55 +2026-08-25 06:42:03 | INFO | train rank=0 | step=7550/100000 progress=7.5% loss=0.095245 lr=9.982233e-06 elapsed=02:46:33 avg_step=1.32s recent_step=1.21s eta=33:59:27 +2026-08-25 06:42:15 | INFO | train rank=0 | step=7560/100000 progress=7.6% loss=0.044022 lr=9.982093e-06 elapsed=02:46:44 avg_step=1.32s recent_step=1.17s eta=33:58:55 +2026-08-25 06:42:28 | INFO | train rank=0 | step=7570/100000 progress=7.6% loss=0.059341 lr=9.981953e-06 elapsed=02:46:58 avg_step=1.32s recent_step=1.35s eta=33:58:45 +2026-08-25 06:42:40 | INFO | train rank=0 | step=7580/100000 progress=7.6% loss=0.049576 lr=9.981813e-06 elapsed=02:47:10 avg_step=1.32s recent_step=1.18s eta=33:58:14 +2026-08-25 06:42:55 | INFO | train rank=0 | step=7590/100000 progress=7.6% loss=0.056885 lr=9.981672e-06 elapsed=02:47:25 avg_step=1.32s recent_step=1.55s eta=33:58:28 +2026-08-25 06:43:11 | INFO | train rank=0 | step=7600/100000 progress=7.6% loss=0.066925 lr=9.981530e-06 elapsed=02:47:40 avg_step=1.32s recent_step=1.53s eta=33:58:40 +2026-08-25 06:43:23 | INFO | train rank=0 | step=7610/100000 progress=7.6% loss=0.088867 lr=9.981388e-06 elapsed=02:47:53 avg_step=1.32s recent_step=1.24s eta=33:58:16 +2026-08-25 06:43:36 | INFO | train rank=0 | step=7620/100000 progress=7.6% loss=0.083984 lr=9.981245e-06 elapsed=02:48:06 avg_step=1.32s recent_step=1.28s eta=33:57:58 +2026-08-25 06:43:50 | INFO | train rank=0 | step=7630/100000 progress=7.6% loss=0.119812 lr=9.981101e-06 elapsed=02:48:20 avg_step=1.32s recent_step=1.39s eta=33:57:53 +2026-08-25 06:44:05 | INFO | train rank=0 | step=7640/100000 progress=7.6% loss=0.091278 lr=9.980957e-06 elapsed=02:48:35 avg_step=1.32s recent_step=1.52s eta=33:58:04 +2026-08-25 06:44:19 | INFO | train rank=0 | step=7650/100000 progress=7.6% loss=0.089172 lr=9.980813e-06 elapsed=02:48:49 avg_step=1.32s recent_step=1.39s eta=33:57:59 +2026-08-25 06:44:30 | INFO | train rank=0 | step=7660/100000 progress=7.7% loss=0.091217 lr=9.980668e-06 elapsed=02:49:00 avg_step=1.32s recent_step=1.16s eta=33:57:26 +2026-08-25 06:44:43 | INFO | train rank=0 | step=7670/100000 progress=7.7% loss=0.076111 lr=9.980523e-06 elapsed=02:49:13 avg_step=1.32s recent_step=1.30s eta=33:57:10 +2026-08-25 06:44:55 | INFO | train rank=0 | step=7680/100000 progress=7.7% loss=0.088654 lr=9.980376e-06 elapsed=02:49:25 avg_step=1.32s recent_step=1.19s eta=33:56:40 +2026-08-25 06:45:08 | INFO | train rank=0 | step=7690/100000 progress=7.7% loss=0.045654 lr=9.980230e-06 elapsed=02:49:38 avg_step=1.32s recent_step=1.29s eta=33:56:23 +2026-08-25 06:45:24 | INFO | train rank=0 | step=7700/100000 progress=7.7% loss=0.097626 lr=9.980083e-06 elapsed=02:49:54 avg_step=1.32s recent_step=1.56s eta=33:56:38 +2026-08-25 06:45:36 | INFO | train rank=0 | step=7710/100000 progress=7.7% loss=0.081177 lr=9.979935e-06 elapsed=02:50:06 avg_step=1.32s recent_step=1.21s eta=33:56:11 +2026-08-25 06:45:50 | INFO | train rank=0 | step=7720/100000 progress=7.7% loss=0.063400 lr=9.979787e-06 elapsed=02:50:20 avg_step=1.32s recent_step=1.39s eta=33:56:05 +2026-08-25 06:46:01 | INFO | train rank=0 | step=7730/100000 progress=7.7% loss=0.094849 lr=9.979638e-06 elapsed=02:50:31 avg_step=1.32s recent_step=1.16s eta=33:55:32 +2026-08-25 06:46:13 | INFO | train rank=0 | step=7740/100000 progress=7.7% loss=0.047409 lr=9.979489e-06 elapsed=02:50:43 avg_step=1.32s recent_step=1.19s eta=33:55:03 +2026-08-25 06:46:29 | INFO | train rank=0 | step=7750/100000 progress=7.8% loss=0.078766 lr=9.979339e-06 elapsed=02:50:59 avg_step=1.32s recent_step=1.56s eta=33:55:18 +2026-08-25 06:46:41 | INFO | train rank=0 | step=7760/100000 progress=7.8% loss=0.070892 lr=9.979188e-06 elapsed=02:51:11 avg_step=1.32s recent_step=1.24s eta=33:54:55 +2026-08-25 06:46:57 | INFO | train rank=0 | step=7770/100000 progress=7.8% loss=0.068542 lr=9.979037e-06 elapsed=02:51:27 avg_step=1.32s recent_step=1.57s eta=33:55:11 +2026-08-25 06:47:09 | INFO | train rank=0 | step=7780/100000 progress=7.8% loss=0.057816 lr=9.978886e-06 elapsed=02:51:39 avg_step=1.32s recent_step=1.21s eta=33:54:44 +2026-08-25 06:47:21 | INFO | train rank=0 | step=7790/100000 progress=7.8% loss=0.087311 lr=9.978734e-06 elapsed=02:51:51 avg_step=1.32s recent_step=1.17s eta=33:54:12 +2026-08-25 06:47:35 | INFO | train rank=0 | step=7800/100000 progress=7.8% loss=0.074066 lr=9.978581e-06 elapsed=02:52:05 avg_step=1.32s recent_step=1.48s eta=33:54:17 +2026-08-25 06:47:47 | INFO | train rank=0 | step=7810/100000 progress=7.8% loss=0.069977 lr=9.978428e-06 elapsed=02:52:17 avg_step=1.32s recent_step=1.17s eta=33:53:45 +2026-08-25 06:47:59 | INFO | train rank=0 | step=7820/100000 progress=7.8% loss=0.054352 lr=9.978274e-06 elapsed=02:52:29 avg_step=1.32s recent_step=1.23s eta=33:53:21 +2026-08-25 06:48:12 | INFO | train rank=0 | step=7830/100000 progress=7.8% loss=0.079468 lr=9.978120e-06 elapsed=02:52:42 avg_step=1.32s recent_step=1.24s eta=33:52:57 +2026-08-25 06:48:30 | INFO | train rank=0 | step=7840/100000 progress=7.8% loss=0.093079 lr=9.977965e-06 elapsed=02:53:00 avg_step=1.32s recent_step=1.82s eta=33:53:42 +2026-08-25 06:48:42 | INFO | train rank=0 | step=7850/100000 progress=7.8% loss=0.072571 lr=9.977810e-06 elapsed=02:53:12 avg_step=1.32s recent_step=1.24s eta=33:53:19 +2026-08-25 06:48:55 | INFO | train rank=0 | step=7860/100000 progress=7.9% loss=0.070557 lr=9.977654e-06 elapsed=02:53:25 avg_step=1.32s recent_step=1.22s eta=33:52:54 +2026-08-25 06:49:08 | INFO | train rank=0 | step=7870/100000 progress=7.9% loss=0.088806 lr=9.977498e-06 elapsed=02:53:38 avg_step=1.32s recent_step=1.34s eta=33:52:42 +2026-08-25 06:49:20 | INFO | train rank=0 | step=7880/100000 progress=7.9% loss=0.130127 lr=9.977341e-06 elapsed=02:53:50 avg_step=1.32s recent_step=1.17s eta=33:52:12 +2026-08-25 06:49:36 | INFO | train rank=0 | step=7890/100000 progress=7.9% loss=0.101715 lr=9.977183e-06 elapsed=02:54:06 avg_step=1.32s recent_step=1.64s eta=33:52:35 +2026-08-25 06:49:49 | INFO | train rank=0 | step=7900/100000 progress=7.9% loss=0.069672 lr=9.977025e-06 elapsed=02:54:19 avg_step=1.32s recent_step=1.27s eta=33:52:15 +2026-08-25 06:50:00 | INFO | train rank=0 | step=7910/100000 progress=7.9% loss=0.152893 lr=9.976866e-06 elapsed=02:54:30 avg_step=1.32s recent_step=1.17s eta=33:51:44 +2026-08-25 06:50:12 | INFO | train rank=0 | step=7920/100000 progress=7.9% loss=0.083710 lr=9.976707e-06 elapsed=02:54:42 avg_step=1.32s recent_step=1.15s eta=33:51:11 +2026-08-25 06:50:27 | INFO | train rank=0 | step=7930/100000 progress=7.9% loss=0.074860 lr=9.976548e-06 elapsed=02:54:56 avg_step=1.32s recent_step=1.46s eta=33:51:13 +2026-08-25 06:50:42 | INFO | train rank=0 | step=7940/100000 progress=7.9% loss=0.086121 lr=9.976387e-06 elapsed=02:55:12 avg_step=1.32s recent_step=1.50s eta=33:51:21 +2026-08-25 06:50:56 | INFO | train rank=0 | step=7950/100000 progress=8.0% loss=0.072357 lr=9.976227e-06 elapsed=02:55:26 avg_step=1.32s recent_step=1.43s eta=33:51:20 +2026-08-25 06:51:08 | INFO | train rank=0 | step=7960/100000 progress=8.0% loss=0.133026 lr=9.976065e-06 elapsed=02:55:38 avg_step=1.32s recent_step=1.19s eta=33:50:50 +2026-08-25 06:51:19 | INFO | train rank=0 | step=7970/100000 progress=8.0% loss=0.061646 lr=9.975903e-06 elapsed=02:55:49 avg_step=1.32s recent_step=1.14s eta=33:50:16 +2026-08-25 06:51:32 | INFO | train rank=0 | step=7980/100000 progress=8.0% loss=0.090698 lr=9.975741e-06 elapsed=02:56:02 avg_step=1.32s recent_step=1.33s eta=33:50:04 +2026-08-25 06:51:44 | INFO | train rank=0 | step=7990/100000 progress=8.0% loss=0.076569 lr=9.975578e-06 elapsed=02:56:14 avg_step=1.32s recent_step=1.17s eta=33:49:32 +2026-08-25 06:52:00 | INFO | train rank=0 | step=8000/100000 progress=8.0% loss=0.105988 lr=9.975415e-06 elapsed=02:56:30 avg_step=1.32s recent_step=1.62s eta=33:49:54 +2026-08-25 06:52:12 | INFO | train rank=0 | step=8010/100000 progress=8.0% loss=0.075470 lr=9.975250e-06 elapsed=02:56:42 avg_step=1.32s recent_step=1.18s eta=33:49:24 +2026-08-25 06:52:24 | INFO | train rank=0 | step=8020/100000 progress=8.0% loss=0.065094 lr=9.975086e-06 elapsed=02:56:54 avg_step=1.32s recent_step=1.17s eta=33:48:53 +2026-08-25 06:52:36 | INFO | train rank=0 | step=8030/100000 progress=8.0% loss=0.069290 lr=9.974921e-06 elapsed=02:57:06 avg_step=1.32s recent_step=1.18s eta=33:48:24 +2026-08-25 06:52:47 | INFO | train rank=0 | step=8040/100000 progress=8.0% loss=0.056702 lr=9.974755e-06 elapsed=02:57:17 avg_step=1.32s recent_step=1.17s eta=33:47:53 +2026-08-25 06:52:59 | INFO | train rank=0 | step=8050/100000 progress=8.1% loss=0.080017 lr=9.974589e-06 elapsed=02:57:29 avg_step=1.32s recent_step=1.14s eta=33:47:18 +2026-08-25 06:53:14 | INFO | train rank=0 | step=8060/100000 progress=8.1% loss=0.125244 lr=9.974422e-06 elapsed=02:57:44 avg_step=1.32s recent_step=1.53s eta=33:47:29 +2026-08-25 06:53:30 | INFO | train rank=0 | step=8070/100000 progress=8.1% loss=0.052811 lr=9.974255e-06 elapsed=02:58:00 avg_step=1.32s recent_step=1.60s eta=33:47:47 +2026-08-25 06:53:42 | INFO | train rank=0 | step=8080/100000 progress=8.1% loss=0.114014 lr=9.974087e-06 elapsed=02:58:12 avg_step=1.32s recent_step=1.20s eta=33:47:20 +2026-08-25 06:53:55 | INFO | train rank=0 | step=8090/100000 progress=8.1% loss=0.063477 lr=9.973919e-06 elapsed=02:58:25 avg_step=1.32s recent_step=1.34s eta=33:47:08 +2026-08-25 06:54:09 | INFO | train rank=0 | step=8100/100000 progress=8.1% loss=0.083649 lr=9.973750e-06 elapsed=02:58:39 avg_step=1.32s recent_step=1.40s eta=33:47:04 +2026-08-25 06:54:22 | INFO | train rank=0 | step=8110/100000 progress=8.1% loss=0.071930 lr=9.973580e-06 elapsed=02:58:52 avg_step=1.32s recent_step=1.24s eta=33:46:41 +2026-08-25 06:54:34 | INFO | train rank=0 | step=8120/100000 progress=8.1% loss=0.047714 lr=9.973410e-06 elapsed=02:59:04 avg_step=1.32s recent_step=1.21s eta=33:46:15 +2026-08-25 06:54:46 | INFO | train rank=0 | step=8130/100000 progress=8.1% loss=0.081329 lr=9.973240e-06 elapsed=02:59:16 avg_step=1.32s recent_step=1.18s eta=33:45:45 +2026-08-25 06:54:58 | INFO | train rank=0 | step=8140/100000 progress=8.1% loss=0.065247 lr=9.973068e-06 elapsed=02:59:28 avg_step=1.32s recent_step=1.22s eta=33:45:20 +2026-08-25 06:55:08 | INFO | train rank=0 | step=8150/100000 progress=8.2% loss=0.055222 lr=9.972897e-06 elapsed=02:59:38 avg_step=1.32s recent_step=1.06s eta=33:44:36 +2026-08-25 06:55:24 | INFO | train rank=0 | step=8160/100000 progress=8.2% loss=0.091248 lr=9.972725e-06 elapsed=02:59:54 avg_step=1.32s recent_step=1.52s eta=33:44:45 +2026-08-25 06:55:39 | INFO | train rank=0 | step=8170/100000 progress=8.2% loss=0.057831 lr=9.972552e-06 elapsed=03:00:09 avg_step=1.32s recent_step=1.55s eta=33:44:58 +2026-08-25 06:55:51 | INFO | train rank=0 | step=8180/100000 progress=8.2% loss=0.096008 lr=9.972379e-06 elapsed=03:00:21 avg_step=1.32s recent_step=1.20s eta=33:44:31 +2026-08-25 06:56:04 | INFO | train rank=0 | step=8190/100000 progress=8.2% loss=0.095673 lr=9.972205e-06 elapsed=03:00:34 avg_step=1.32s recent_step=1.27s eta=33:44:12 +2026-08-25 06:56:16 | INFO | train rank=0 | step=8200/100000 progress=8.2% loss=0.076355 lr=9.972030e-06 elapsed=03:00:46 avg_step=1.32s recent_step=1.22s eta=33:43:47 +2026-08-25 06:56:32 | INFO | train rank=0 | step=8210/100000 progress=8.2% loss=0.070862 lr=9.971855e-06 elapsed=03:01:02 avg_step=1.32s recent_step=1.62s eta=33:44:07 +2026-08-25 06:56:45 | INFO | train rank=0 | step=8220/100000 progress=8.2% loss=0.106018 lr=9.971680e-06 elapsed=03:01:15 avg_step=1.32s recent_step=1.25s eta=33:43:46 +2026-08-25 06:57:00 | INFO | train rank=0 | step=8230/100000 progress=8.2% loss=0.066650 lr=9.971504e-06 elapsed=03:01:30 avg_step=1.32s recent_step=1.49s eta=33:43:52 +2026-08-25 06:57:13 | INFO | train rank=0 | step=8240/100000 progress=8.2% loss=0.099548 lr=9.971327e-06 elapsed=03:01:42 avg_step=1.32s recent_step=1.28s eta=33:43:33 +2026-08-25 06:57:27 | INFO | train rank=0 | step=8250/100000 progress=8.2% loss=0.114014 lr=9.971150e-06 elapsed=03:01:57 avg_step=1.32s recent_step=1.43s eta=33:43:32 +2026-08-25 06:57:39 | INFO | train rank=0 | step=8260/100000 progress=8.3% loss=0.169861 lr=9.970973e-06 elapsed=03:02:09 avg_step=1.32s recent_step=1.18s eta=33:43:03 +2026-08-25 06:57:51 | INFO | train rank=0 | step=8270/100000 progress=8.3% loss=0.096039 lr=9.970794e-06 elapsed=03:02:21 avg_step=1.32s recent_step=1.20s eta=33:42:36 +2026-08-25 06:58:03 | INFO | train rank=0 | step=8280/100000 progress=8.3% loss=0.110718 lr=9.970616e-06 elapsed=03:02:32 avg_step=1.32s recent_step=1.19s eta=33:42:08 +2026-08-25 06:58:19 | INFO | train rank=0 | step=8290/100000 progress=8.3% loss=0.074829 lr=9.970436e-06 elapsed=03:02:49 avg_step=1.32s recent_step=1.64s eta=33:42:29 +2026-08-25 06:58:31 | INFO | train rank=0 | step=8300/100000 progress=8.3% loss=0.113098 lr=9.970257e-06 elapsed=03:03:01 avg_step=1.32s recent_step=1.22s eta=33:42:05 +2026-08-25 06:58:49 | INFO | train rank=0 | step=8310/100000 progress=8.3% loss=0.092255 lr=9.970076e-06 elapsed=03:03:19 avg_step=1.32s recent_step=1.78s eta=33:42:42 +2026-08-25 06:59:01 | INFO | train rank=0 | step=8320/100000 progress=8.3% loss=0.102203 lr=9.969895e-06 elapsed=03:03:31 avg_step=1.32s recent_step=1.19s eta=33:42:14 +2026-08-25 06:59:12 | INFO | train rank=0 | step=8330/100000 progress=8.3% loss=0.098328 lr=9.969714e-06 elapsed=03:03:42 avg_step=1.32s recent_step=1.16s eta=33:41:43 +2026-08-25 06:59:24 | INFO | train rank=0 | step=8340/100000 progress=8.3% loss=0.081604 lr=9.969532e-06 elapsed=03:03:54 avg_step=1.32s recent_step=1.15s eta=33:41:11 +2026-08-25 06:59:36 | INFO | train rank=0 | step=8350/100000 progress=8.3% loss=0.070770 lr=9.969349e-06 elapsed=03:04:06 avg_step=1.32s recent_step=1.25s eta=33:40:50 +2026-08-25 06:59:49 | INFO | train rank=0 | step=8360/100000 progress=8.4% loss=0.055878 lr=9.969166e-06 elapsed=03:04:19 avg_step=1.32s recent_step=1.26s eta=33:40:29 +2026-08-25 07:00:04 | INFO | train rank=0 | step=8370/100000 progress=8.4% loss=0.078156 lr=9.968983e-06 elapsed=03:04:34 avg_step=1.32s recent_step=1.53s eta=33:40:39 +2026-08-25 07:00:16 | INFO | train rank=0 | step=8380/100000 progress=8.4% loss=0.100861 lr=9.968799e-06 elapsed=03:04:46 avg_step=1.32s recent_step=1.20s eta=33:40:12 +2026-08-25 07:00:28 | INFO | train rank=0 | step=8390/100000 progress=8.4% loss=0.103943 lr=9.968614e-06 elapsed=03:04:58 avg_step=1.32s recent_step=1.21s eta=33:39:46 +2026-08-25 07:00:41 | INFO | train rank=0 | step=8400/100000 progress=8.4% loss=0.050598 lr=9.968429e-06 elapsed=03:05:11 avg_step=1.32s recent_step=1.29s eta=33:39:29 +2026-08-25 07:00:53 | INFO | train rank=0 | step=8410/100000 progress=8.4% loss=0.128174 lr=9.968243e-06 elapsed=03:05:23 avg_step=1.32s recent_step=1.16s eta=33:38:58 +2026-08-25 07:01:05 | INFO | train rank=0 | step=8420/100000 progress=8.4% loss=0.092682 lr=9.968057e-06 elapsed=03:05:35 avg_step=1.32s recent_step=1.19s eta=33:38:30 +2026-08-25 07:01:17 | INFO | train rank=0 | step=8430/100000 progress=8.4% loss=0.125916 lr=9.967870e-06 elapsed=03:05:47 avg_step=1.32s recent_step=1.22s eta=33:38:06 +2026-08-25 07:01:29 | INFO | train rank=0 | step=8440/100000 progress=8.4% loss=0.084564 lr=9.967682e-06 elapsed=03:05:59 avg_step=1.32s recent_step=1.23s eta=33:37:44 +2026-08-25 07:01:44 | INFO | train rank=0 | step=8450/100000 progress=8.5% loss=0.082306 lr=9.967494e-06 elapsed=03:06:14 avg_step=1.32s recent_step=1.50s eta=33:37:50 +2026-08-25 07:01:56 | INFO | train rank=0 | step=8460/100000 progress=8.5% loss=0.122803 lr=9.967306e-06 elapsed=03:06:26 avg_step=1.32s recent_step=1.16s eta=33:37:19 +2026-08-25 07:02:10 | INFO | train rank=0 | step=8470/100000 progress=8.5% loss=0.047150 lr=9.967117e-06 elapsed=03:06:40 avg_step=1.32s recent_step=1.45s eta=33:37:20 +2026-08-25 07:02:25 | INFO | train rank=0 | step=8480/100000 progress=8.5% loss=0.106018 lr=9.966927e-06 elapsed=03:06:55 avg_step=1.32s recent_step=1.43s eta=33:37:18 +2026-08-25 07:02:36 | INFO | train rank=0 | step=8490/100000 progress=8.5% loss=0.052795 lr=9.966737e-06 elapsed=03:07:06 avg_step=1.32s recent_step=1.17s eta=33:36:49 +2026-08-25 07:02:51 | INFO | train rank=0 | step=8500/100000 progress=8.5% loss=0.097656 lr=9.966546e-06 elapsed=03:07:21 avg_step=1.32s recent_step=1.42s eta=33:36:46 +2026-08-25 07:03:03 | INFO | train rank=0 | step=8510/100000 progress=8.5% loss=0.062012 lr=9.966355e-06 elapsed=03:07:33 avg_step=1.32s recent_step=1.23s eta=33:36:24 +2026-08-25 07:03:18 | INFO | train rank=0 | step=8520/100000 progress=8.5% loss=0.089050 lr=9.966163e-06 elapsed=03:07:48 avg_step=1.32s recent_step=1.51s eta=33:36:30 +2026-08-25 07:03:33 | INFO | train rank=0 | step=8530/100000 progress=8.5% loss=0.117981 lr=9.965971e-06 elapsed=03:08:03 avg_step=1.32s recent_step=1.50s eta=33:36:36 +2026-08-25 07:03:44 | INFO | train rank=0 | step=8540/100000 progress=8.5% loss=0.089996 lr=9.965778e-06 elapsed=03:08:14 avg_step=1.32s recent_step=1.11s eta=33:36:00 +2026-08-25 07:03:58 | INFO | train rank=0 | step=8550/100000 progress=8.6% loss=0.062012 lr=9.965585e-06 elapsed=03:08:28 avg_step=1.32s recent_step=1.38s eta=33:35:54 +2026-08-25 07:04:17 | INFO | train rank=0 | step=8560/100000 progress=8.6% loss=0.067749 lr=9.965391e-06 elapsed=03:08:46 avg_step=1.32s recent_step=1.85s eta=33:36:36 +2026-08-25 07:04:28 | INFO | train rank=0 | step=8570/100000 progress=8.6% loss=0.075592 lr=9.965196e-06 elapsed=03:08:58 avg_step=1.32s recent_step=1.19s eta=33:36:09 +2026-08-25 07:04:41 | INFO | train rank=0 | step=8580/100000 progress=8.6% loss=0.072021 lr=9.965001e-06 elapsed=03:09:11 avg_step=1.32s recent_step=1.23s eta=33:35:46 +2026-08-25 07:04:54 | INFO | train rank=0 | step=8590/100000 progress=8.6% loss=0.060867 lr=9.964806e-06 elapsed=03:09:24 avg_step=1.32s recent_step=1.32s eta=33:35:32 +2026-08-25 07:05:08 | INFO | train rank=0 | step=8600/100000 progress=8.6% loss=0.122131 lr=9.964610e-06 elapsed=03:09:38 avg_step=1.32s recent_step=1.43s eta=33:35:30 +2026-08-25 07:05:23 | INFO | train rank=0 | step=8610/100000 progress=8.6% loss=0.061432 lr=9.964413e-06 elapsed=03:09:53 avg_step=1.32s recent_step=1.52s eta=33:35:38 +2026-08-25 07:05:37 | INFO | train rank=0 | step=8620/100000 progress=8.6% loss=0.078369 lr=9.964216e-06 elapsed=03:10:07 avg_step=1.32s recent_step=1.38s eta=33:35:31 +2026-08-25 07:05:49 | INFO | train rank=0 | step=8630/100000 progress=8.6% loss=0.093262 lr=9.964018e-06 elapsed=03:10:19 avg_step=1.32s recent_step=1.19s eta=33:35:04 +2026-08-25 07:06:03 | INFO | train rank=0 | step=8640/100000 progress=8.6% loss=0.092896 lr=9.963820e-06 elapsed=03:10:33 avg_step=1.32s recent_step=1.35s eta=33:34:53 +2026-08-25 07:06:15 | INFO | train rank=0 | step=8650/100000 progress=8.6% loss=0.080170 lr=9.963621e-06 elapsed=03:10:45 avg_step=1.32s recent_step=1.21s eta=33:34:28 +2026-08-25 07:06:27 | INFO | train rank=0 | step=8660/100000 progress=8.7% loss=0.053696 lr=9.963422e-06 elapsed=03:10:56 avg_step=1.32s recent_step=1.19s eta=33:34:00 +2026-08-25 07:06:38 | INFO | train rank=0 | step=8670/100000 progress=8.7% loss=0.080902 lr=9.963222e-06 elapsed=03:11:08 avg_step=1.32s recent_step=1.15s eta=33:33:29 +2026-08-25 07:06:50 | INFO | train rank=0 | step=8680/100000 progress=8.7% loss=0.082062 lr=9.963021e-06 elapsed=03:11:20 avg_step=1.32s recent_step=1.22s eta=33:33:05 +2026-08-25 07:07:03 | INFO | train rank=0 | step=8690/100000 progress=8.7% loss=0.049042 lr=9.962820e-06 elapsed=03:11:33 avg_step=1.32s recent_step=1.31s eta=33:32:51 +2026-08-25 07:07:15 | INFO | train rank=0 | step=8700/100000 progress=8.7% loss=0.105347 lr=9.962619e-06 elapsed=03:11:45 avg_step=1.32s recent_step=1.18s eta=33:32:23 +2026-08-25 07:07:31 | INFO | train rank=0 | step=8710/100000 progress=8.7% loss=0.079346 lr=9.962417e-06 elapsed=03:12:01 avg_step=1.32s recent_step=1.61s eta=33:32:39 +2026-08-25 07:07:43 | INFO | train rank=0 | step=8720/100000 progress=8.7% loss=0.064362 lr=9.962214e-06 elapsed=03:12:13 avg_step=1.32s recent_step=1.21s eta=33:32:14 +2026-08-25 07:07:55 | INFO | train rank=0 | step=8730/100000 progress=8.7% loss=0.058914 lr=9.962011e-06 elapsed=03:12:25 avg_step=1.32s recent_step=1.16s eta=33:31:43 +2026-08-25 07:08:11 | INFO | train rank=0 | step=8740/100000 progress=8.7% loss=0.054779 lr=9.961807e-06 elapsed=03:12:41 avg_step=1.32s recent_step=1.57s eta=33:31:56 +2026-08-25 07:08:27 | INFO | train rank=0 | step=8750/100000 progress=8.8% loss=0.101318 lr=9.961603e-06 elapsed=03:12:57 avg_step=1.32s recent_step=1.66s eta=33:32:18 +2026-08-25 07:08:40 | INFO | train rank=0 | step=8760/100000 progress=8.8% loss=0.083099 lr=9.961398e-06 elapsed=03:13:10 avg_step=1.32s recent_step=1.26s eta=33:31:58 +2026-08-25 07:08:52 | INFO | train rank=0 | step=8770/100000 progress=8.8% loss=0.060471 lr=9.961193e-06 elapsed=03:13:21 avg_step=1.32s recent_step=1.17s eta=33:31:29 +2026-08-25 07:09:08 | INFO | train rank=0 | step=8780/100000 progress=8.8% loss=0.090912 lr=9.960987e-06 elapsed=03:13:38 avg_step=1.32s recent_step=1.63s eta=33:31:48 +2026-08-25 07:09:20 | INFO | train rank=0 | step=8790/100000 progress=8.8% loss=0.075439 lr=9.960780e-06 elapsed=03:13:50 avg_step=1.32s recent_step=1.19s eta=33:31:22 +2026-08-25 07:09:34 | INFO | train rank=0 | step=8800/100000 progress=8.8% loss=0.061707 lr=9.960574e-06 elapsed=03:14:04 avg_step=1.32s recent_step=1.40s eta=33:31:16 +2026-08-25 07:09:46 | INFO | train rank=0 | step=8810/100000 progress=8.8% loss=0.078735 lr=9.960366e-06 elapsed=03:14:16 avg_step=1.32s recent_step=1.21s eta=33:30:51 +2026-08-25 07:09:59 | INFO | train rank=0 | step=8820/100000 progress=8.8% loss=0.054001 lr=9.960158e-06 elapsed=03:14:29 avg_step=1.32s recent_step=1.27s eta=33:30:32 +2026-08-25 07:10:12 | INFO | train rank=0 | step=8830/100000 progress=8.8% loss=0.108521 lr=9.959949e-06 elapsed=03:14:42 avg_step=1.32s recent_step=1.39s eta=33:30:26 +2026-08-25 07:10:25 | INFO | train rank=0 | step=8840/100000 progress=8.8% loss=0.063248 lr=9.959740e-06 elapsed=03:14:55 avg_step=1.32s recent_step=1.28s eta=33:30:08 +2026-08-25 07:10:37 | INFO | train rank=0 | step=8850/100000 progress=8.8% loss=0.067322 lr=9.959531e-06 elapsed=03:15:07 avg_step=1.32s recent_step=1.22s eta=33:29:44 +2026-08-25 07:10:52 | INFO | train rank=0 | step=8860/100000 progress=8.9% loss=0.073074 lr=9.959320e-06 elapsed=03:15:22 avg_step=1.32s recent_step=1.44s eta=33:29:43 +2026-08-25 07:11:07 | INFO | train rank=0 | step=8870/100000 progress=8.9% loss=0.160889 lr=9.959110e-06 elapsed=03:15:36 avg_step=1.32s recent_step=1.47s eta=33:29:45 +2026-08-25 07:11:18 | INFO | train rank=0 | step=8880/100000 progress=8.9% loss=0.053452 lr=9.958898e-06 elapsed=03:15:48 avg_step=1.32s recent_step=1.15s eta=33:29:14 +2026-08-25 07:11:30 | INFO | train rank=0 | step=8890/100000 progress=8.9% loss=0.082520 lr=9.958686e-06 elapsed=03:16:00 avg_step=1.32s recent_step=1.20s eta=33:28:49 +2026-08-25 07:11:44 | INFO | train rank=0 | step=8900/100000 progress=8.9% loss=0.053864 lr=9.958474e-06 elapsed=03:16:14 avg_step=1.32s recent_step=1.36s eta=33:28:39 +2026-08-25 07:11:59 | INFO | train rank=0 | step=8910/100000 progress=8.9% loss=0.054733 lr=9.958261e-06 elapsed=03:16:29 avg_step=1.32s recent_step=1.55s eta=33:28:49 +2026-08-25 07:12:15 | INFO | train rank=0 | step=8920/100000 progress=8.9% loss=0.077026 lr=9.958048e-06 elapsed=03:16:44 avg_step=1.32s recent_step=1.53s eta=33:28:57 +2026-08-25 07:12:26 | INFO | train rank=0 | step=8930/100000 progress=8.9% loss=0.071350 lr=9.957834e-06 elapsed=03:16:56 avg_step=1.32s recent_step=1.16s eta=33:28:27 +2026-08-25 07:12:40 | INFO | train rank=0 | step=8940/100000 progress=8.9% loss=0.073044 lr=9.957619e-06 elapsed=03:17:10 avg_step=1.32s recent_step=1.35s eta=33:28:17 +2026-08-25 07:12:53 | INFO | train rank=0 | step=8950/100000 progress=8.9% loss=0.078186 lr=9.957404e-06 elapsed=03:17:23 avg_step=1.32s recent_step=1.35s eta=33:28:06 +2026-08-25 07:13:05 | INFO | train rank=0 | step=8960/100000 progress=9.0% loss=0.092346 lr=9.957188e-06 elapsed=03:17:35 avg_step=1.32s recent_step=1.23s eta=33:27:43 +2026-08-25 07:13:17 | INFO | train rank=0 | step=8970/100000 progress=9.0% loss=0.084290 lr=9.956972e-06 elapsed=03:17:47 avg_step=1.32s recent_step=1.20s eta=33:27:18 +2026-08-25 07:13:31 | INFO | train rank=0 | step=8980/100000 progress=9.0% loss=0.091309 lr=9.956755e-06 elapsed=03:18:01 avg_step=1.32s recent_step=1.38s eta=33:27:10 +2026-08-25 07:13:47 | INFO | train rank=0 | step=8990/100000 progress=9.0% loss=0.065582 lr=9.956538e-06 elapsed=03:18:17 avg_step=1.32s recent_step=1.60s eta=33:27:25 +2026-08-25 07:14:00 | INFO | train rank=0 | step=9000/100000 progress=9.0% loss=0.086273 lr=9.956320e-06 elapsed=03:18:29 avg_step=1.32s recent_step=1.23s eta=33:27:02 +2026-08-25 07:14:11 | INFO | train rank=0 | step=9010/100000 progress=9.0% loss=0.071106 lr=9.956102e-06 elapsed=03:18:41 avg_step=1.32s recent_step=1.15s eta=33:26:32 +2026-08-25 07:14:24 | INFO | train rank=0 | step=9020/100000 progress=9.0% loss=0.065186 lr=9.955883e-06 elapsed=03:18:53 avg_step=1.32s recent_step=1.25s eta=33:26:11 +2026-08-25 07:14:35 | INFO | train rank=0 | step=9030/100000 progress=9.0% loss=0.076050 lr=9.955664e-06 elapsed=03:19:05 avg_step=1.32s recent_step=1.18s eta=33:25:43 +2026-08-25 07:14:49 | INFO | train rank=0 | step=9040/100000 progress=9.0% loss=0.092407 lr=9.955444e-06 elapsed=03:19:19 avg_step=1.32s recent_step=1.38s eta=33:25:35 +2026-08-25 07:15:01 | INFO | train rank=0 | step=9050/100000 progress=9.0% loss=0.071838 lr=9.955223e-06 elapsed=03:19:31 avg_step=1.32s recent_step=1.20s eta=33:25:10 +2026-08-25 07:15:12 | INFO | train rank=0 | step=9060/100000 progress=9.1% loss=0.041367 lr=9.955002e-06 elapsed=03:19:42 avg_step=1.32s recent_step=1.10s eta=33:24:35 +2026-08-25 07:15:24 | INFO | train rank=0 | step=9070/100000 progress=9.1% loss=0.139771 lr=9.954781e-06 elapsed=03:19:54 avg_step=1.32s recent_step=1.23s eta=33:24:12 +2026-08-25 07:15:38 | INFO | train rank=0 | step=9080/100000 progress=9.1% loss=0.067535 lr=9.954558e-06 elapsed=03:20:08 avg_step=1.32s recent_step=1.33s eta=33:24:00 +2026-08-25 07:15:52 | INFO | train rank=0 | step=9090/100000 progress=9.1% loss=0.076538 lr=9.954336e-06 elapsed=03:20:22 avg_step=1.32s recent_step=1.44s eta=33:23:59 +2026-08-25 07:16:04 | INFO | train rank=0 | step=9100/100000 progress=9.1% loss=0.061493 lr=9.954112e-06 elapsed=03:20:34 avg_step=1.32s recent_step=1.19s eta=33:23:33 +2026-08-25 07:16:19 | INFO | train rank=0 | step=9110/100000 progress=9.1% loss=0.063202 lr=9.953889e-06 elapsed=03:20:49 avg_step=1.32s recent_step=1.49s eta=33:23:37 +2026-08-25 07:16:31 | INFO | train rank=0 | step=9120/100000 progress=9.1% loss=0.126099 lr=9.953664e-06 elapsed=03:21:01 avg_step=1.32s recent_step=1.16s eta=33:23:07 +2026-08-25 07:16:44 | INFO | train rank=0 | step=9130/100000 progress=9.1% loss=0.097290 lr=9.953440e-06 elapsed=03:21:14 avg_step=1.32s recent_step=1.30s eta=33:22:52 +2026-08-25 07:16:58 | INFO | train rank=0 | step=9140/100000 progress=9.1% loss=0.099640 lr=9.953214e-06 elapsed=03:21:27 avg_step=1.32s recent_step=1.38s eta=33:22:45 +2026-08-25 07:17:13 | INFO | train rank=0 | step=9150/100000 progress=9.2% loss=0.071594 lr=9.952988e-06 elapsed=03:21:43 avg_step=1.32s recent_step=1.54s eta=33:22:53 +2026-08-25 07:17:28 | INFO | train rank=0 | step=9160/100000 progress=9.2% loss=0.081299 lr=9.952762e-06 elapsed=03:21:58 avg_step=1.32s recent_step=1.54s eta=33:23:01 +2026-08-25 07:17:40 | INFO | train rank=0 | step=9170/100000 progress=9.2% loss=0.085297 lr=9.952535e-06 elapsed=03:22:10 avg_step=1.32s recent_step=1.16s eta=33:22:32 +2026-08-25 07:17:54 | INFO | train rank=0 | step=9180/100000 progress=9.2% loss=0.091980 lr=9.952307e-06 elapsed=03:22:24 avg_step=1.32s recent_step=1.43s eta=33:22:30 +2026-08-25 07:18:10 | INFO | train rank=0 | step=9190/100000 progress=9.2% loss=0.085663 lr=9.952079e-06 elapsed=03:22:40 avg_step=1.32s recent_step=1.58s eta=33:22:42 +2026-08-25 07:18:25 | INFO | train rank=0 | step=9200/100000 progress=9.2% loss=0.099243 lr=9.951850e-06 elapsed=03:22:55 avg_step=1.32s recent_step=1.50s eta=33:22:46 +2026-08-25 07:18:39 | INFO | train rank=0 | step=9210/100000 progress=9.2% loss=0.068420 lr=9.951621e-06 elapsed=03:23:09 avg_step=1.32s recent_step=1.40s eta=33:22:40 +2026-08-25 07:18:51 | INFO | train rank=0 | step=9220/100000 progress=9.2% loss=0.063950 lr=9.951391e-06 elapsed=03:23:21 avg_step=1.32s recent_step=1.17s eta=33:22:12 +2026-08-25 07:19:03 | INFO | train rank=0 | step=9230/100000 progress=9.2% loss=0.101593 lr=9.951161e-06 elapsed=03:23:33 avg_step=1.32s recent_step=1.27s eta=33:21:53 +2026-08-25 07:19:15 | INFO | train rank=0 | step=9240/100000 progress=9.2% loss=0.114655 lr=9.950930e-06 elapsed=03:23:45 avg_step=1.32s recent_step=1.19s eta=33:21:27 +2026-08-25 07:19:30 | INFO | train rank=0 | step=9250/100000 progress=9.2% loss=0.092255 lr=9.950699e-06 elapsed=03:23:59 avg_step=1.32s recent_step=1.42s eta=33:21:23 +2026-08-25 07:19:42 | INFO | train rank=0 | step=9260/100000 progress=9.3% loss=0.072021 lr=9.950467e-06 elapsed=03:24:12 avg_step=1.32s recent_step=1.23s eta=33:21:01 +2026-08-25 07:19:59 | INFO | train rank=0 | step=9270/100000 progress=9.3% loss=0.056839 lr=9.950235e-06 elapsed=03:24:29 avg_step=1.32s recent_step=1.75s eta=33:21:30 +2026-08-25 07:20:11 | INFO | train rank=0 | step=9280/100000 progress=9.3% loss=0.087311 lr=9.950002e-06 elapsed=03:24:41 avg_step=1.32s recent_step=1.17s eta=33:21:01 +2026-08-25 07:20:25 | INFO | train rank=0 | step=9290/100000 progress=9.3% loss=0.070374 lr=9.949768e-06 elapsed=03:24:55 avg_step=1.32s recent_step=1.43s eta=33:20:59 +2026-08-25 07:20:40 | INFO | train rank=0 | step=9300/100000 progress=9.3% loss=0.061066 lr=9.949534e-06 elapsed=03:25:10 avg_step=1.32s recent_step=1.48s eta=33:21:01 +2026-08-25 07:20:52 | INFO | train rank=0 | step=9310/100000 progress=9.3% loss=0.063507 lr=9.949300e-06 elapsed=03:25:22 avg_step=1.32s recent_step=1.20s eta=33:20:36 +2026-08-25 07:21:04 | INFO | train rank=0 | step=9320/100000 progress=9.3% loss=0.060516 lr=9.949064e-06 elapsed=03:25:34 avg_step=1.32s recent_step=1.20s eta=33:20:11 +2026-08-25 07:21:16 | INFO | train rank=0 | step=9330/100000 progress=9.3% loss=0.061523 lr=9.948829e-06 elapsed=03:25:46 avg_step=1.32s recent_step=1.20s eta=33:19:45 +2026-08-25 07:21:28 | INFO | train rank=0 | step=9340/100000 progress=9.3% loss=0.087524 lr=9.948593e-06 elapsed=03:25:58 avg_step=1.32s recent_step=1.16s eta=33:19:16 +2026-08-25 07:21:40 | INFO | train rank=0 | step=9350/100000 progress=9.3% loss=0.060669 lr=9.948356e-06 elapsed=03:26:10 avg_step=1.32s recent_step=1.18s eta=33:18:50 +2026-08-25 07:21:53 | INFO | train rank=0 | step=9360/100000 progress=9.4% loss=0.098633 lr=9.948118e-06 elapsed=03:26:23 avg_step=1.32s recent_step=1.33s eta=33:18:37 +2026-08-25 07:22:08 | INFO | train rank=0 | step=9370/100000 progress=9.4% loss=0.078232 lr=9.947881e-06 elapsed=03:26:38 avg_step=1.32s recent_step=1.55s eta=33:18:45 +2026-08-25 07:22:20 | INFO | train rank=0 | step=9380/100000 progress=9.4% loss=0.070374 lr=9.947642e-06 elapsed=03:26:50 avg_step=1.32s recent_step=1.16s eta=33:18:16 +2026-08-25 07:22:34 | INFO | train rank=0 | step=9390/100000 progress=9.4% loss=0.091614 lr=9.947403e-06 elapsed=03:27:04 avg_step=1.32s recent_step=1.40s eta=33:18:11 +2026-08-25 07:22:48 | INFO | train rank=0 | step=9400/100000 progress=9.4% loss=0.113770 lr=9.947164e-06 elapsed=03:27:18 avg_step=1.32s recent_step=1.36s eta=33:18:01 +2026-08-25 07:22:59 | INFO | train rank=0 | step=9410/100000 progress=9.4% loss=0.041168 lr=9.946924e-06 elapsed=03:27:29 avg_step=1.32s recent_step=1.16s eta=33:17:32 +2026-08-25 07:23:11 | INFO | train rank=0 | step=9420/100000 progress=9.4% loss=0.065887 lr=9.946683e-06 elapsed=03:27:41 avg_step=1.32s recent_step=1.15s eta=33:17:02 +2026-08-25 07:23:25 | INFO | train rank=0 | step=9430/100000 progress=9.4% loss=0.064148 lr=9.946442e-06 elapsed=03:27:55 avg_step=1.32s recent_step=1.46s eta=33:17:03 +2026-08-25 07:23:41 | INFO | train rank=0 | step=9440/100000 progress=9.4% loss=0.067932 lr=9.946201e-06 elapsed=03:28:10 avg_step=1.32s recent_step=1.51s eta=33:17:08 +2026-08-25 07:23:55 | INFO | train rank=0 | step=9450/100000 progress=9.4% loss=0.066315 lr=9.945958e-06 elapsed=03:28:25 avg_step=1.32s recent_step=1.45s eta=33:17:06 +2026-08-25 07:24:07 | INFO | train rank=0 | step=9460/100000 progress=9.5% loss=0.079163 lr=9.945716e-06 elapsed=03:28:37 avg_step=1.32s recent_step=1.20s eta=33:16:41 +2026-08-25 07:24:18 | INFO | train rank=0 | step=9470/100000 progress=9.5% loss=0.068115 lr=9.945472e-06 elapsed=03:28:48 avg_step=1.32s recent_step=1.14s eta=33:16:10 +2026-08-25 07:24:34 | INFO | train rank=0 | step=9480/100000 progress=9.5% loss=0.087616 lr=9.945229e-06 elapsed=03:29:04 avg_step=1.32s recent_step=1.59s eta=33:16:23 +2026-08-25 07:24:51 | INFO | train rank=0 | step=9490/100000 progress=9.5% loss=0.103638 lr=9.944984e-06 elapsed=03:29:21 avg_step=1.32s recent_step=1.69s eta=33:16:45 +2026-08-25 07:25:06 | INFO | train rank=0 | step=9500/100000 progress=9.5% loss=0.147461 lr=9.944739e-06 elapsed=03:29:36 avg_step=1.32s recent_step=1.51s eta=33:16:49 +2026-08-25 07:25:19 | INFO | train rank=0 | step=9510/100000 progress=9.5% loss=0.109436 lr=9.944494e-06 elapsed=03:29:49 avg_step=1.32s recent_step=1.26s eta=33:16:30 +2026-08-25 07:25:31 | INFO | train rank=0 | step=9520/100000 progress=9.5% loss=0.068375 lr=9.944248e-06 elapsed=03:30:01 avg_step=1.32s recent_step=1.24s eta=33:16:08 +2026-08-25 07:25:43 | INFO | train rank=0 | step=9530/100000 progress=9.5% loss=0.071167 lr=9.944001e-06 elapsed=03:30:13 avg_step=1.32s recent_step=1.19s eta=33:15:42 +2026-08-25 07:26:01 | INFO | train rank=0 | step=9540/100000 progress=9.5% loss=0.042343 lr=9.943754e-06 elapsed=03:30:31 avg_step=1.32s recent_step=1.75s eta=33:16:09 +2026-08-25 07:26:12 | INFO | train rank=0 | step=9550/100000 progress=9.6% loss=0.066833 lr=9.943507e-06 elapsed=03:30:42 avg_step=1.32s recent_step=1.15s eta=33:15:40 +2026-08-25 07:26:24 | INFO | train rank=0 | step=9560/100000 progress=9.6% loss=0.067596 lr=9.943259e-06 elapsed=03:30:54 avg_step=1.32s recent_step=1.20s eta=33:15:15 +2026-08-25 07:26:39 | INFO | train rank=0 | step=9570/100000 progress=9.6% loss=0.066467 lr=9.943010e-06 elapsed=03:31:09 avg_step=1.32s recent_step=1.50s eta=33:15:18 +2026-08-25 07:26:53 | INFO | train rank=0 | step=9580/100000 progress=9.6% loss=0.123596 lr=9.942761e-06 elapsed=03:31:23 avg_step=1.32s recent_step=1.41s eta=33:15:13 +2026-08-25 07:27:05 | INFO | train rank=0 | step=9590/100000 progress=9.6% loss=0.112366 lr=9.942511e-06 elapsed=03:31:35 avg_step=1.32s recent_step=1.14s eta=33:14:42 +2026-08-25 07:27:16 | INFO | train rank=0 | step=9600/100000 progress=9.6% loss=0.102661 lr=9.942261e-06 elapsed=03:31:46 avg_step=1.32s recent_step=1.18s eta=33:14:15 +2026-08-25 07:27:29 | INFO | train rank=0 | step=9610/100000 progress=9.6% loss=0.050079 lr=9.942010e-06 elapsed=03:31:59 avg_step=1.32s recent_step=1.26s eta=33:13:56 +2026-08-25 07:27:43 | INFO | train rank=0 | step=9620/100000 progress=9.6% loss=0.105225 lr=9.941759e-06 elapsed=03:32:13 avg_step=1.32s recent_step=1.41s eta=33:13:51 +2026-08-25 07:27:59 | INFO | train rank=0 | step=9630/100000 progress=9.6% loss=0.099243 lr=9.941507e-06 elapsed=03:32:28 avg_step=1.32s recent_step=1.54s eta=33:13:58 +2026-08-25 07:28:11 | INFO | train rank=0 | step=9640/100000 progress=9.6% loss=0.117554 lr=9.941254e-06 elapsed=03:32:40 avg_step=1.32s recent_step=1.20s eta=33:13:34 +2026-08-25 07:28:22 | INFO | train rank=0 | step=9650/100000 progress=9.7% loss=0.071808 lr=9.941001e-06 elapsed=03:32:52 avg_step=1.32s recent_step=1.18s eta=33:13:07 +2026-08-25 07:28:34 | INFO | train rank=0 | step=9660/100000 progress=9.7% loss=0.056030 lr=9.940748e-06 elapsed=03:33:04 avg_step=1.32s recent_step=1.18s eta=33:12:40 +2026-08-25 07:28:47 | INFO | train rank=0 | step=9670/100000 progress=9.7% loss=0.152100 lr=9.940494e-06 elapsed=03:33:17 avg_step=1.32s recent_step=1.33s eta=33:12:27 +2026-08-25 07:28:59 | INFO | train rank=0 | step=9680/100000 progress=9.7% loss=0.145447 lr=9.940239e-06 elapsed=03:33:29 avg_step=1.32s recent_step=1.17s eta=33:11:59 +2026-08-25 07:29:15 | INFO | train rank=0 | step=9690/100000 progress=9.7% loss=0.066956 lr=9.939984e-06 elapsed=03:33:44 avg_step=1.32s recent_step=1.55s eta=33:12:07 +2026-08-25 07:29:26 | INFO | train rank=0 | step=9700/100000 progress=9.7% loss=0.061554 lr=9.939728e-06 elapsed=03:33:56 avg_step=1.32s recent_step=1.15s eta=33:11:38 +2026-08-25 07:29:40 | INFO | train rank=0 | step=9710/100000 progress=9.7% loss=0.131409 lr=9.939472e-06 elapsed=03:34:10 avg_step=1.32s recent_step=1.42s eta=33:11:34 +2026-08-25 07:29:51 | INFO | train rank=0 | step=9720/100000 progress=9.7% loss=0.067932 lr=9.939215e-06 elapsed=03:34:21 avg_step=1.32s recent_step=1.12s eta=33:11:01 +2026-08-25 07:30:08 | INFO | train rank=0 | step=9730/100000 progress=9.7% loss=0.072388 lr=9.938958e-06 elapsed=03:34:37 avg_step=1.32s recent_step=1.61s eta=33:11:15 +2026-08-25 07:30:19 | INFO | train rank=0 | step=9740/100000 progress=9.7% loss=0.084610 lr=9.938700e-06 elapsed=03:34:49 avg_step=1.32s recent_step=1.16s eta=33:10:47 +2026-08-25 07:30:34 | INFO | train rank=0 | step=9750/100000 progress=9.8% loss=0.103577 lr=9.938442e-06 elapsed=03:35:04 avg_step=1.32s recent_step=1.46s eta=33:10:46 +2026-08-25 07:30:45 | INFO | train rank=0 | step=9760/100000 progress=9.8% loss=0.073669 lr=9.938183e-06 elapsed=03:35:15 avg_step=1.32s recent_step=1.14s eta=33:10:16 +2026-08-25 07:31:03 | INFO | train rank=0 | step=9770/100000 progress=9.8% loss=0.095978 lr=9.937923e-06 elapsed=03:35:33 avg_step=1.32s recent_step=1.80s eta=33:10:47 +2026-08-25 07:31:16 | INFO | train rank=0 | step=9780/100000 progress=9.8% loss=0.057098 lr=9.937663e-06 elapsed=03:35:46 avg_step=1.32s recent_step=1.24s eta=33:10:27 +2026-08-25 07:31:28 | INFO | train rank=0 | step=9790/100000 progress=9.8% loss=0.070297 lr=9.937403e-06 elapsed=03:35:58 avg_step=1.32s recent_step=1.26s eta=33:10:08 +2026-08-25 07:31:44 | INFO | train rank=0 | step=9800/100000 progress=9.8% loss=0.138977 lr=9.937142e-06 elapsed=03:36:14 avg_step=1.32s recent_step=1.57s eta=33:10:17 +2026-08-25 07:31:56 | INFO | train rank=0 | step=9810/100000 progress=9.8% loss=0.065765 lr=9.936880e-06 elapsed=03:36:26 avg_step=1.32s recent_step=1.20s eta=33:09:53 +2026-08-25 07:32:08 | INFO | train rank=0 | step=9820/100000 progress=9.8% loss=0.109619 lr=9.936618e-06 elapsed=03:36:37 avg_step=1.32s recent_step=1.15s eta=33:09:24 +2026-08-25 07:32:21 | INFO | train rank=0 | step=9830/100000 progress=9.8% loss=0.064117 lr=9.936355e-06 elapsed=03:36:51 avg_step=1.32s recent_step=1.31s eta=33:09:10 +2026-08-25 07:32:36 | INFO | train rank=0 | step=9840/100000 progress=9.8% loss=0.110535 lr=9.936092e-06 elapsed=03:37:06 avg_step=1.32s recent_step=1.50s eta=33:09:13 +2026-08-25 07:32:47 | INFO | train rank=0 | step=9850/100000 progress=9.8% loss=0.071228 lr=9.935828e-06 elapsed=03:37:17 avg_step=1.32s recent_step=1.15s eta=33:08:44 +2026-08-25 07:32:59 | INFO | train rank=0 | step=9860/100000 progress=9.9% loss=0.085754 lr=9.935564e-06 elapsed=03:37:29 avg_step=1.32s recent_step=1.14s eta=33:08:14 +2026-08-25 07:33:11 | INFO | train rank=0 | step=9870/100000 progress=9.9% loss=0.097778 lr=9.935299e-06 elapsed=03:37:41 avg_step=1.32s recent_step=1.26s eta=33:07:55 +2026-08-25 07:33:23 | INFO | train rank=0 | step=9880/100000 progress=9.9% loss=0.060837 lr=9.935034e-06 elapsed=03:37:53 avg_step=1.32s recent_step=1.16s eta=33:07:27 +2026-08-25 07:33:38 | INFO | train rank=0 | step=9890/100000 progress=9.9% loss=0.057068 lr=9.934768e-06 elapsed=03:38:07 avg_step=1.32s recent_step=1.47s eta=33:07:27 +2026-08-25 07:33:53 | INFO | train rank=0 | step=9900/100000 progress=9.9% loss=0.083069 lr=9.934501e-06 elapsed=03:38:23 avg_step=1.32s recent_step=1.59s eta=33:07:38 +2026-08-25 07:34:05 | INFO | train rank=0 | step=9910/100000 progress=9.9% loss=0.061050 lr=9.934234e-06 elapsed=03:38:35 avg_step=1.32s recent_step=1.18s eta=33:07:11 +2026-08-25 07:34:16 | INFO | train rank=0 | step=9920/100000 progress=9.9% loss=0.073364 lr=9.933966e-06 elapsed=03:38:46 avg_step=1.32s recent_step=1.11s eta=33:06:39 +2026-08-25 07:34:31 | INFO | train rank=0 | step=9930/100000 progress=9.9% loss=0.071198 lr=9.933698e-06 elapsed=03:39:01 avg_step=1.32s recent_step=1.49s eta=33:06:41 +2026-08-25 07:34:45 | INFO | train rank=0 | step=9940/100000 progress=9.9% loss=0.059662 lr=9.933430e-06 elapsed=03:39:15 avg_step=1.32s recent_step=1.38s eta=33:06:33 +2026-08-25 07:35:00 | INFO | train rank=0 | step=9950/100000 progress=10.0% loss=0.037750 lr=9.933161e-06 elapsed=03:39:30 avg_step=1.32s recent_step=1.54s eta=33:06:39 +2026-08-25 07:35:14 | INFO | train rank=0 | step=9960/100000 progress=10.0% loss=0.089539 lr=9.932891e-06 elapsed=03:39:44 avg_step=1.32s recent_step=1.37s eta=33:06:29 +2026-08-25 07:35:28 | INFO | train rank=0 | step=9970/100000 progress=10.0% loss=0.095642 lr=9.932621e-06 elapsed=03:39:58 avg_step=1.32s recent_step=1.36s eta=33:06:19 +2026-08-25 07:35:40 | INFO | train rank=0 | step=9980/100000 progress=10.0% loss=0.115906 lr=9.932350e-06 elapsed=03:40:10 avg_step=1.32s recent_step=1.20s eta=33:05:55 +2026-08-25 07:35:55 | INFO | train rank=0 | step=9990/100000 progress=10.0% loss=0.061539 lr=9.932078e-06 elapsed=03:40:25 avg_step=1.32s recent_step=1.51s eta=33:05:58 +2026-08-25 07:36:06 | INFO | train rank=0 | step=10000/100000 progress=10.0% loss=0.115845 lr=9.931807e-06 elapsed=03:40:36 avg_step=1.32s recent_step=1.18s eta=33:05:31 +2026-08-25 07:36:19 | INFO | train rank=0 | step=10010/100000 progress=10.0% loss=0.097168 lr=9.931534e-06 elapsed=03:40:49 avg_step=1.32s recent_step=1.28s eta=33:05:14 +2026-08-25 07:36:32 | INFO | train rank=0 | step=10020/100000 progress=10.0% loss=0.124451 lr=9.931261e-06 elapsed=03:41:02 avg_step=1.32s recent_step=1.25s eta=33:04:54 +2026-08-25 07:36:45 | INFO | train rank=0 | step=10030/100000 progress=10.0% loss=0.087891 lr=9.930988e-06 elapsed=03:41:15 avg_step=1.32s recent_step=1.37s eta=33:04:45 +2026-08-25 07:36:58 | INFO | train rank=0 | step=10040/100000 progress=10.0% loss=0.048386 lr=9.930714e-06 elapsed=03:41:28 avg_step=1.32s recent_step=1.24s eta=33:04:24 +2026-08-25 07:37:17 | INFO | train rank=0 | step=10050/100000 progress=10.1% loss=0.042450 lr=9.930439e-06 elapsed=03:41:46 avg_step=1.32s recent_step=1.87s eta=33:05:00 +2026-08-25 07:37:28 | INFO | train rank=0 | step=10060/100000 progress=10.1% loss=0.099579 lr=9.930164e-06 elapsed=03:41:58 avg_step=1.32s recent_step=1.19s eta=33:04:35 +2026-08-25 07:37:41 | INFO | train rank=0 | step=10070/100000 progress=10.1% loss=0.068008 lr=9.929888e-06 elapsed=03:42:10 avg_step=1.32s recent_step=1.21s eta=33:04:11 +2026-08-25 07:37:52 | INFO | train rank=0 | step=10080/100000 progress=10.1% loss=0.080139 lr=9.929612e-06 elapsed=03:42:22 avg_step=1.32s recent_step=1.15s eta=33:03:43 +2026-08-25 07:38:06 | INFO | train rank=0 | step=10090/100000 progress=10.1% loss=0.056046 lr=9.929335e-06 elapsed=03:42:36 avg_step=1.32s recent_step=1.40s eta=33:03:36 +2026-08-25 07:38:18 | INFO | train rank=0 | step=10100/100000 progress=10.1% loss=0.174194 lr=9.929058e-06 elapsed=03:42:47 avg_step=1.32s recent_step=1.15s eta=33:03:08 +2026-08-25 07:38:30 | INFO | train rank=0 | step=10110/100000 progress=10.1% loss=0.055298 lr=9.928780e-06 elapsed=03:43:00 avg_step=1.32s recent_step=1.22s eta=33:02:45 +2026-08-25 07:38:41 | INFO | train rank=0 | step=10120/100000 progress=10.1% loss=0.049911 lr=9.928502e-06 elapsed=03:43:11 avg_step=1.32s recent_step=1.17s eta=33:02:18 +2026-08-25 07:38:57 | INFO | train rank=0 | step=10130/100000 progress=10.1% loss=0.092712 lr=9.928223e-06 elapsed=03:43:26 avg_step=1.32s recent_step=1.51s eta=33:02:22 +2026-08-25 07:39:08 | INFO | train rank=0 | step=10140/100000 progress=10.1% loss=0.134949 lr=9.927944e-06 elapsed=03:43:38 avg_step=1.32s recent_step=1.18s eta=33:01:55 +2026-08-25 07:39:21 | INFO | train rank=0 | step=10150/100000 progress=10.2% loss=0.062973 lr=9.927664e-06 elapsed=03:43:51 avg_step=1.32s recent_step=1.24s eta=33:01:34 +2026-08-25 07:39:33 | INFO | train rank=0 | step=10160/100000 progress=10.2% loss=0.073639 lr=9.927383e-06 elapsed=03:44:03 avg_step=1.32s recent_step=1.21s eta=33:01:11 +2026-08-25 07:39:51 | INFO | train rank=0 | step=10170/100000 progress=10.2% loss=0.053192 lr=9.927102e-06 elapsed=03:44:21 avg_step=1.32s recent_step=1.78s eta=33:01:39 +2026-08-25 07:40:03 | INFO | train rank=0 | step=10180/100000 progress=10.2% loss=0.100769 lr=9.926820e-06 elapsed=03:44:33 avg_step=1.32s recent_step=1.24s eta=33:01:18 +2026-08-25 07:40:15 | INFO | train rank=0 | step=10190/100000 progress=10.2% loss=0.059662 lr=9.926538e-06 elapsed=03:44:45 avg_step=1.32s recent_step=1.19s eta=33:00:53 +2026-08-25 07:40:29 | INFO | train rank=0 | step=10200/100000 progress=10.2% loss=0.069336 lr=9.926256e-06 elapsed=03:44:59 avg_step=1.32s recent_step=1.43s eta=33:00:50 +2026-08-25 07:40:47 | INFO | train rank=0 | step=10210/100000 progress=10.2% loss=0.045227 lr=9.925972e-06 elapsed=03:45:16 avg_step=1.32s recent_step=1.72s eta=33:01:12 +2026-08-25 07:40:58 | INFO | train rank=0 | step=10220/100000 progress=10.2% loss=0.060242 lr=9.925689e-06 elapsed=03:45:28 avg_step=1.32s recent_step=1.18s eta=33:00:45 +2026-08-25 07:41:10 | INFO | train rank=0 | step=10230/100000 progress=10.2% loss=0.055710 lr=9.925404e-06 elapsed=03:45:40 avg_step=1.32s recent_step=1.20s eta=33:00:21 +2026-08-25 07:41:25 | INFO | train rank=0 | step=10240/100000 progress=10.2% loss=0.072968 lr=9.925120e-06 elapsed=03:45:55 avg_step=1.32s recent_step=1.48s eta=33:00:22 +2026-08-25 07:41:37 | INFO | train rank=0 | step=10250/100000 progress=10.2% loss=0.068756 lr=9.924834e-06 elapsed=03:46:07 avg_step=1.32s recent_step=1.21s eta=32:59:59 +2026-08-25 07:41:49 | INFO | train rank=0 | step=10260/100000 progress=10.3% loss=0.120728 lr=9.924548e-06 elapsed=03:46:19 avg_step=1.32s recent_step=1.16s eta=32:59:31 +2026-08-25 07:42:01 | INFO | train rank=0 | step=10270/100000 progress=10.3% loss=0.078247 lr=9.924262e-06 elapsed=03:46:31 avg_step=1.32s recent_step=1.23s eta=32:59:09 +2026-08-25 07:42:13 | INFO | train rank=0 | step=10280/100000 progress=10.3% loss=0.073883 lr=9.923975e-06 elapsed=03:46:43 avg_step=1.32s recent_step=1.18s eta=32:58:44 +2026-08-25 07:42:26 | INFO | train rank=0 | step=10290/100000 progress=10.3% loss=0.058685 lr=9.923687e-06 elapsed=03:46:56 avg_step=1.32s recent_step=1.28s eta=32:58:26 +2026-08-25 07:42:38 | INFO | train rank=0 | step=10300/100000 progress=10.3% loss=0.065948 lr=9.923399e-06 elapsed=03:47:08 avg_step=1.32s recent_step=1.20s eta=32:58:03 +2026-08-25 07:42:50 | INFO | train rank=0 | step=10310/100000 progress=10.3% loss=0.107300 lr=9.923111e-06 elapsed=03:47:20 avg_step=1.32s recent_step=1.28s eta=32:57:46 +2026-08-25 07:43:03 | INFO | train rank=0 | step=10320/100000 progress=10.3% loss=0.059402 lr=9.922822e-06 elapsed=03:47:32 avg_step=1.32s recent_step=1.22s eta=32:57:23 +2026-08-25 07:43:16 | INFO | train rank=0 | step=10330/100000 progress=10.3% loss=0.099884 lr=9.922532e-06 elapsed=03:47:46 avg_step=1.32s recent_step=1.30s eta=32:57:08 +2026-08-25 07:43:31 | INFO | train rank=0 | step=10340/100000 progress=10.3% loss=0.062469 lr=9.922242e-06 elapsed=03:48:01 avg_step=1.32s recent_step=1.57s eta=32:57:16 +2026-08-25 07:43:43 | INFO | train rank=0 | step=10350/100000 progress=10.3% loss=0.083740 lr=9.921951e-06 elapsed=03:48:13 avg_step=1.32s recent_step=1.21s eta=32:56:53 +2026-08-25 07:44:02 | INFO | train rank=0 | step=10360/100000 progress=10.4% loss=0.091644 lr=9.921660e-06 elapsed=03:48:32 avg_step=1.32s recent_step=1.87s eta=32:57:27 +2026-08-25 07:44:14 | INFO | train rank=0 | step=10370/100000 progress=10.4% loss=0.078735 lr=9.921368e-06 elapsed=03:48:44 avg_step=1.32s recent_step=1.15s eta=32:56:59 +2026-08-25 07:44:30 | INFO | train rank=0 | step=10380/100000 progress=10.4% loss=0.056992 lr=9.921076e-06 elapsed=03:49:00 avg_step=1.32s recent_step=1.62s eta=32:57:11 +2026-08-25 07:44:46 | INFO | train rank=0 | step=10390/100000 progress=10.4% loss=0.083618 lr=9.920783e-06 elapsed=03:49:15 avg_step=1.32s recent_step=1.58s eta=32:57:20 +2026-08-25 07:45:00 | INFO | train rank=0 | step=10400/100000 progress=10.4% loss=0.073486 lr=9.920489e-06 elapsed=03:49:30 avg_step=1.32s recent_step=1.40s eta=32:57:13 +2026-08-25 07:45:11 | INFO | train rank=0 | step=10410/100000 progress=10.4% loss=0.062408 lr=9.920195e-06 elapsed=03:49:41 avg_step=1.32s recent_step=1.13s eta=32:56:44 +2026-08-25 07:45:23 | INFO | train rank=0 | step=10420/100000 progress=10.4% loss=0.098938 lr=9.919901e-06 elapsed=03:49:53 avg_step=1.32s recent_step=1.22s eta=32:56:21 +2026-08-25 07:45:38 | INFO | train rank=0 | step=10430/100000 progress=10.4% loss=0.090912 lr=9.919606e-06 elapsed=03:50:07 avg_step=1.32s recent_step=1.45s eta=32:56:19 +2026-08-25 07:45:49 | INFO | train rank=0 | step=10440/100000 progress=10.4% loss=0.083435 lr=9.919310e-06 elapsed=03:50:19 avg_step=1.32s recent_step=1.17s eta=32:55:52 +2026-08-25 07:46:04 | INFO | train rank=0 | step=10450/100000 progress=10.4% loss=0.055786 lr=9.919014e-06 elapsed=03:50:34 avg_step=1.32s recent_step=1.48s eta=32:55:53 +2026-08-25 07:46:16 | INFO | train rank=0 | step=10460/100000 progress=10.5% loss=0.041519 lr=9.918717e-06 elapsed=03:50:46 avg_step=1.32s recent_step=1.19s eta=32:55:28 +2026-08-25 07:46:28 | INFO | train rank=0 | step=10470/100000 progress=10.5% loss=0.077362 lr=9.918420e-06 elapsed=03:50:58 avg_step=1.32s recent_step=1.20s eta=32:55:04 +2026-08-25 07:46:40 | INFO | train rank=0 | step=10480/100000 progress=10.5% loss=0.037842 lr=9.918122e-06 elapsed=03:51:10 avg_step=1.32s recent_step=1.20s eta=32:54:41 +2026-08-25 07:46:52 | INFO | train rank=0 | step=10490/100000 progress=10.5% loss=0.081635 lr=9.917824e-06 elapsed=03:51:22 avg_step=1.32s recent_step=1.18s eta=32:54:15 +2026-08-25 07:47:07 | INFO | train rank=0 | step=10500/100000 progress=10.5% loss=0.106476 lr=9.917525e-06 elapsed=03:51:37 avg_step=1.32s recent_step=1.52s eta=32:54:18 +2026-08-25 07:47:19 | INFO | train rank=0 | step=10510/100000 progress=10.5% loss=0.102112 lr=9.917226e-06 elapsed=03:51:49 avg_step=1.32s recent_step=1.18s eta=32:53:53 +2026-08-25 07:47:36 | INFO | train rank=0 | step=10520/100000 progress=10.5% loss=0.057892 lr=9.916926e-06 elapsed=03:52:05 avg_step=1.32s recent_step=1.67s eta=32:54:09 +2026-08-25 07:47:48 | INFO | train rank=0 | step=10530/100000 progress=10.5% loss=0.049164 lr=9.916626e-06 elapsed=03:52:18 avg_step=1.32s recent_step=1.22s eta=32:53:47 +2026-08-25 07:48:01 | INFO | train rank=0 | step=10540/100000 progress=10.5% loss=0.054474 lr=9.916325e-06 elapsed=03:52:31 avg_step=1.32s recent_step=1.30s eta=32:53:32 +2026-08-25 07:48:13 | INFO | train rank=0 | step=10550/100000 progress=10.5% loss=0.055695 lr=9.916023e-06 elapsed=03:52:43 avg_step=1.32s recent_step=1.20s eta=32:53:08 +2026-08-25 07:48:25 | INFO | train rank=0 | step=10560/100000 progress=10.6% loss=0.061462 lr=9.915721e-06 elapsed=03:52:55 avg_step=1.32s recent_step=1.20s eta=32:52:45 +2026-08-25 07:48:39 | INFO | train rank=0 | step=10570/100000 progress=10.6% loss=0.052948 lr=9.915419e-06 elapsed=03:53:08 avg_step=1.32s recent_step=1.38s eta=32:52:37 +2026-08-25 07:48:53 | INFO | train rank=0 | step=10580/100000 progress=10.6% loss=0.068634 lr=9.915116e-06 elapsed=03:53:23 avg_step=1.32s recent_step=1.47s eta=32:52:36 +2026-08-25 07:49:10 | INFO | train rank=0 | step=10590/100000 progress=10.6% loss=0.085083 lr=9.914812e-06 elapsed=03:53:39 avg_step=1.32s recent_step=1.63s eta=32:52:48 +2026-08-25 07:49:21 | INFO | train rank=0 | step=10600/100000 progress=10.6% loss=0.074951 lr=9.914508e-06 elapsed=03:53:51 avg_step=1.32s recent_step=1.18s eta=32:52:23 +2026-08-25 07:49:36 | INFO | train rank=0 | step=10610/100000 progress=10.6% loss=0.081207 lr=9.914203e-06 elapsed=03:54:06 avg_step=1.32s recent_step=1.48s eta=32:52:23 +2026-08-25 07:49:52 | INFO | train rank=0 | step=10620/100000 progress=10.6% loss=0.080109 lr=9.913898e-06 elapsed=03:54:22 avg_step=1.32s recent_step=1.55s eta=32:52:29 +2026-08-25 07:50:06 | INFO | train rank=0 | step=10630/100000 progress=10.6% loss=0.067169 lr=9.913592e-06 elapsed=03:54:36 avg_step=1.32s recent_step=1.46s eta=32:52:27 +2026-08-25 07:50:18 | INFO | train rank=0 | step=10640/100000 progress=10.6% loss=0.084717 lr=9.913286e-06 elapsed=03:54:48 avg_step=1.32s recent_step=1.18s eta=32:52:02 +2026-08-25 07:50:30 | INFO | train rank=0 | step=10650/100000 progress=10.7% loss=0.077332 lr=9.912979e-06 elapsed=03:55:00 avg_step=1.32s recent_step=1.22s eta=32:51:39 +2026-08-25 07:50:47 | INFO | train rank=0 | step=10660/100000 progress=10.7% loss=0.062714 lr=9.912671e-06 elapsed=03:55:16 avg_step=1.32s recent_step=1.63s eta=32:51:51 +2026-08-25 07:50:58 | INFO | train rank=0 | step=10670/100000 progress=10.7% loss=0.063690 lr=9.912363e-06 elapsed=03:55:28 avg_step=1.32s recent_step=1.16s eta=32:51:24 +2026-08-25 07:51:10 | INFO | train rank=0 | step=10680/100000 progress=10.7% loss=0.119934 lr=9.912055e-06 elapsed=03:55:40 avg_step=1.32s recent_step=1.17s eta=32:50:58 +2026-08-25 07:51:25 | INFO | train rank=0 | step=10690/100000 progress=10.7% loss=0.066650 lr=9.911746e-06 elapsed=03:55:55 avg_step=1.32s recent_step=1.50s eta=32:50:59 +2026-08-25 07:51:37 | INFO | train rank=0 | step=10700/100000 progress=10.7% loss=0.086334 lr=9.911436e-06 elapsed=03:56:06 avg_step=1.32s recent_step=1.18s eta=32:50:34 +2026-08-25 07:51:50 | INFO | train rank=0 | step=10710/100000 progress=10.7% loss=0.064987 lr=9.911126e-06 elapsed=03:56:20 avg_step=1.32s recent_step=1.34s eta=32:50:22 +2026-08-25 07:52:04 | INFO | train rank=0 | step=10720/100000 progress=10.7% loss=0.093567 lr=9.910816e-06 elapsed=03:56:34 avg_step=1.32s recent_step=1.44s eta=32:50:19 +2026-08-25 07:52:16 | INFO | train rank=0 | step=10730/100000 progress=10.7% loss=0.090240 lr=9.910504e-06 elapsed=03:56:46 avg_step=1.32s recent_step=1.16s eta=32:49:51 +2026-08-25 07:52:30 | INFO | train rank=0 | step=10740/100000 progress=10.7% loss=0.066101 lr=9.910193e-06 elapsed=03:57:00 avg_step=1.32s recent_step=1.44s eta=32:49:48 +2026-08-25 07:52:42 | INFO | train rank=0 | step=10750/100000 progress=10.8% loss=0.064911 lr=9.909880e-06 elapsed=03:57:12 avg_step=1.32s recent_step=1.17s eta=32:49:22 +2026-08-25 07:52:54 | INFO | train rank=0 | step=10760/100000 progress=10.8% loss=0.069412 lr=9.909568e-06 elapsed=03:57:24 avg_step=1.32s recent_step=1.22s eta=32:49:00 +2026-08-25 07:53:07 | INFO | train rank=0 | step=10770/100000 progress=10.8% loss=0.051270 lr=9.909254e-06 elapsed=03:57:37 avg_step=1.32s recent_step=1.31s eta=32:48:45 +2026-08-25 07:53:20 | INFO | train rank=0 | step=10780/100000 progress=10.8% loss=0.053665 lr=9.908940e-06 elapsed=03:57:50 avg_step=1.32s recent_step=1.30s eta=32:48:30 +2026-08-25 07:53:34 | INFO | train rank=0 | step=10790/100000 progress=10.8% loss=0.086945 lr=9.908626e-06 elapsed=03:58:04 avg_step=1.32s recent_step=1.38s eta=32:48:22 +2026-08-25 07:53:48 | INFO | train rank=0 | step=10800/100000 progress=10.8% loss=0.057831 lr=9.908311e-06 elapsed=03:58:18 avg_step=1.32s recent_step=1.43s eta=32:48:17 +2026-08-25 07:54:00 | INFO | train rank=0 | step=10810/100000 progress=10.8% loss=0.048401 lr=9.907996e-06 elapsed=03:58:30 avg_step=1.32s recent_step=1.17s eta=32:47:51 +2026-08-25 07:54:15 | INFO | train rank=0 | step=10820/100000 progress=10.8% loss=0.077545 lr=9.907680e-06 elapsed=03:58:45 avg_step=1.32s recent_step=1.47s eta=32:47:50 +2026-08-25 07:54:27 | INFO | train rank=0 | step=10830/100000 progress=10.8% loss=0.087097 lr=9.907363e-06 elapsed=03:58:57 avg_step=1.32s recent_step=1.19s eta=32:47:26 +2026-08-25 07:54:40 | INFO | train rank=0 | step=10840/100000 progress=10.8% loss=0.086761 lr=9.907046e-06 elapsed=03:59:10 avg_step=1.32s recent_step=1.35s eta=32:47:15 +2026-08-25 07:54:53 | INFO | train rank=0 | step=10850/100000 progress=10.8% loss=0.071991 lr=9.906728e-06 elapsed=03:59:23 avg_step=1.32s recent_step=1.27s eta=32:46:57 +2026-08-25 07:55:07 | INFO | train rank=0 | step=10860/100000 progress=10.9% loss=0.091492 lr=9.906410e-06 elapsed=03:59:37 avg_step=1.32s recent_step=1.38s eta=32:46:49 +2026-08-25 07:55:23 | INFO | train rank=0 | step=10870/100000 progress=10.9% loss=0.039444 lr=9.906092e-06 elapsed=03:59:53 avg_step=1.32s recent_step=1.61s eta=32:46:59 +2026-08-25 07:55:34 | INFO | train rank=0 | step=10880/100000 progress=10.9% loss=0.062042 lr=9.905772e-06 elapsed=04:00:04 avg_step=1.32s recent_step=1.16s eta=32:46:32 +2026-08-25 07:55:49 | INFO | train rank=0 | step=10890/100000 progress=10.9% loss=0.061127 lr=9.905453e-06 elapsed=04:00:19 avg_step=1.32s recent_step=1.45s eta=32:46:29 +2026-08-25 07:56:01 | INFO | train rank=0 | step=10900/100000 progress=10.9% loss=0.095703 lr=9.905132e-06 elapsed=04:00:31 avg_step=1.32s recent_step=1.24s eta=32:46:09 +2026-08-25 07:56:13 | INFO | train rank=0 | step=10910/100000 progress=10.9% loss=0.081573 lr=9.904811e-06 elapsed=04:00:43 avg_step=1.32s recent_step=1.18s eta=32:45:44 +2026-08-25 07:56:26 | INFO | train rank=0 | step=10920/100000 progress=10.9% loss=0.071014 lr=9.904490e-06 elapsed=04:00:56 avg_step=1.32s recent_step=1.28s eta=32:45:27 +2026-08-25 07:56:38 | INFO | train rank=0 | step=10930/100000 progress=10.9% loss=0.069733 lr=9.904168e-06 elapsed=04:01:08 avg_step=1.32s recent_step=1.25s eta=32:45:08 +2026-08-25 07:56:53 | INFO | train rank=0 | step=10940/100000 progress=10.9% loss=0.074615 lr=9.903846e-06 elapsed=04:01:23 avg_step=1.32s recent_step=1.49s eta=32:45:08 +2026-08-25 07:57:05 | INFO | train rank=0 | step=10950/100000 progress=10.9% loss=0.054443 lr=9.903523e-06 elapsed=04:01:35 avg_step=1.32s recent_step=1.18s eta=32:44:44 +2026-08-25 07:57:17 | INFO | train rank=0 | step=10960/100000 progress=11.0% loss=0.095215 lr=9.903199e-06 elapsed=04:01:47 avg_step=1.32s recent_step=1.22s eta=32:44:22 +2026-08-25 07:57:29 | INFO | train rank=0 | step=10970/100000 progress=11.0% loss=0.063416 lr=9.902875e-06 elapsed=04:01:59 avg_step=1.32s recent_step=1.17s eta=32:43:56 +2026-08-25 07:57:43 | INFO | train rank=0 | step=10980/100000 progress=11.0% loss=0.050659 lr=9.902551e-06 elapsed=04:02:13 avg_step=1.32s recent_step=1.45s eta=32:43:53 +2026-08-25 07:57:56 | INFO | train rank=0 | step=10990/100000 progress=11.0% loss=0.058929 lr=9.902225e-06 elapsed=04:02:26 avg_step=1.32s recent_step=1.24s eta=32:43:33 +2026-08-25 07:58:13 | INFO | train rank=0 | step=11000/100000 progress=11.0% loss=0.070374 lr=9.901900e-06 elapsed=04:02:43 avg_step=1.32s recent_step=1.72s eta=32:43:52 +2026-08-25 07:58:25 | INFO | train rank=0 | step=11010/100000 progress=11.0% loss=0.059723 lr=9.901574e-06 elapsed=04:02:55 avg_step=1.32s recent_step=1.17s eta=32:43:26 +2026-08-25 07:58:40 | INFO | train rank=0 | step=11020/100000 progress=11.0% loss=0.108032 lr=9.901247e-06 elapsed=04:03:10 avg_step=1.32s recent_step=1.48s eta=32:43:26 +2026-08-25 07:58:55 | INFO | train rank=0 | step=11030/100000 progress=11.0% loss=0.048630 lr=9.900920e-06 elapsed=04:03:24 avg_step=1.32s recent_step=1.49s eta=32:43:26 +2026-08-25 07:59:07 | INFO | train rank=0 | step=11040/100000 progress=11.0% loss=0.080200 lr=9.900592e-06 elapsed=04:03:37 avg_step=1.32s recent_step=1.25s eta=32:43:07 +2026-08-25 07:59:18 | INFO | train rank=0 | step=11050/100000 progress=11.1% loss=0.068726 lr=9.900263e-06 elapsed=04:03:48 avg_step=1.32s recent_step=1.08s eta=32:42:34 +2026-08-25 07:59:30 | INFO | train rank=0 | step=11060/100000 progress=11.1% loss=0.077972 lr=9.899935e-06 elapsed=04:04:00 avg_step=1.32s recent_step=1.17s eta=32:42:09 +2026-08-25 07:59:41 | INFO | train rank=0 | step=11070/100000 progress=11.1% loss=0.035400 lr=9.899605e-06 elapsed=04:04:11 avg_step=1.32s recent_step=1.17s eta=32:41:43 +2026-08-25 07:59:57 | INFO | train rank=0 | step=11080/100000 progress=11.1% loss=0.067993 lr=9.899275e-06 elapsed=04:04:27 avg_step=1.32s recent_step=1.53s eta=32:41:46 +2026-08-25 08:00:10 | INFO | train rank=0 | step=11090/100000 progress=11.1% loss=0.051651 lr=9.898945e-06 elapsed=04:04:40 avg_step=1.32s recent_step=1.30s eta=32:41:32 +2026-08-25 08:00:25 | INFO | train rank=0 | step=11100/100000 progress=11.1% loss=0.065552 lr=9.898614e-06 elapsed=04:04:55 avg_step=1.32s recent_step=1.51s eta=32:41:33 +2026-08-25 08:00:40 | INFO | train rank=0 | step=11110/100000 progress=11.1% loss=0.068909 lr=9.898282e-06 elapsed=04:05:10 avg_step=1.32s recent_step=1.54s eta=32:41:37 +2026-08-25 08:00:55 | INFO | train rank=0 | step=11120/100000 progress=11.1% loss=0.129272 lr=9.897950e-06 elapsed=04:05:25 avg_step=1.32s recent_step=1.48s eta=32:41:37 +2026-08-25 08:01:06 | INFO | train rank=0 | step=11130/100000 progress=11.1% loss=0.082733 lr=9.897617e-06 elapsed=04:05:36 avg_step=1.32s recent_step=1.13s eta=32:41:08 +2026-08-25 08:01:21 | INFO | train rank=0 | step=11140/100000 progress=11.1% loss=0.049988 lr=9.897284e-06 elapsed=04:05:51 avg_step=1.32s recent_step=1.47s eta=32:41:06 +2026-08-25 08:01:32 | INFO | train rank=0 | step=11150/100000 progress=11.2% loss=0.066132 lr=9.896951e-06 elapsed=04:06:02 avg_step=1.32s recent_step=1.14s eta=32:40:39 +2026-08-25 08:01:50 | INFO | train rank=0 | step=11160/100000 progress=11.2% loss=0.064896 lr=9.896616e-06 elapsed=04:06:20 avg_step=1.32s recent_step=1.74s eta=32:40:58 +2026-08-25 08:02:03 | INFO | train rank=0 | step=11170/100000 progress=11.2% loss=0.092773 lr=9.896282e-06 elapsed=04:06:33 avg_step=1.32s recent_step=1.29s eta=32:40:42 +2026-08-25 08:02:14 | INFO | train rank=0 | step=11180/100000 progress=11.2% loss=0.096252 lr=9.895946e-06 elapsed=04:06:44 avg_step=1.32s recent_step=1.11s eta=32:40:12 +2026-08-25 08:02:26 | INFO | train rank=0 | step=11190/100000 progress=11.2% loss=0.061371 lr=9.895610e-06 elapsed=04:06:56 avg_step=1.32s recent_step=1.19s eta=32:39:47 +2026-08-25 08:02:37 | INFO | train rank=0 | step=11200/100000 progress=11.2% loss=0.057861 lr=9.895274e-06 elapsed=04:07:07 avg_step=1.32s recent_step=1.15s eta=32:39:20 +2026-08-25 08:02:49 | INFO | train rank=0 | step=11210/100000 progress=11.2% loss=0.058380 lr=9.894937e-06 elapsed=04:07:18 avg_step=1.32s recent_step=1.15s eta=32:38:53 +2026-08-25 08:03:01 | INFO | train rank=0 | step=11220/100000 progress=11.2% loss=0.055695 lr=9.894600e-06 elapsed=04:07:30 avg_step=1.32s recent_step=1.20s eta=32:38:30 +2026-08-25 08:03:12 | INFO | train rank=0 | step=11230/100000 progress=11.2% loss=0.081116 lr=9.894262e-06 elapsed=04:07:42 avg_step=1.32s recent_step=1.14s eta=32:38:03 +2026-08-25 08:03:28 | INFO | train rank=0 | step=11240/100000 progress=11.2% loss=0.069397 lr=9.893923e-06 elapsed=04:07:57 avg_step=1.32s recent_step=1.56s eta=32:38:08 +2026-08-25 08:03:43 | INFO | train rank=0 | step=11250/100000 progress=11.2% loss=0.055908 lr=9.893584e-06 elapsed=04:08:13 avg_step=1.32s recent_step=1.55s eta=32:38:12 +2026-08-25 08:03:56 | INFO | train rank=0 | step=11260/100000 progress=11.3% loss=0.082031 lr=9.893245e-06 elapsed=04:08:26 avg_step=1.32s recent_step=1.32s eta=32:37:58 +2026-08-25 08:04:12 | INFO | train rank=0 | step=11270/100000 progress=11.3% loss=0.049408 lr=9.892905e-06 elapsed=04:08:41 avg_step=1.32s recent_step=1.54s eta=32:38:02 +2026-08-25 08:04:26 | INFO | train rank=0 | step=11280/100000 progress=11.3% loss=0.070221 lr=9.892564e-06 elapsed=04:08:56 avg_step=1.32s recent_step=1.42s eta=32:37:57 +2026-08-25 08:04:37 | INFO | train rank=0 | step=11290/100000 progress=11.3% loss=0.082520 lr=9.892223e-06 elapsed=04:09:07 avg_step=1.32s recent_step=1.16s eta=32:37:31 +2026-08-25 08:04:51 | INFO | train rank=0 | step=11300/100000 progress=11.3% loss=0.066055 lr=9.891881e-06 elapsed=04:09:21 avg_step=1.32s recent_step=1.32s eta=32:37:17 +2026-08-25 08:05:03 | INFO | train rank=0 | step=11310/100000 progress=11.3% loss=0.065521 lr=9.891539e-06 elapsed=04:09:33 avg_step=1.32s recent_step=1.29s eta=32:37:01 +2026-08-25 08:05:18 | INFO | train rank=0 | step=11320/100000 progress=11.3% loss=0.073761 lr=9.891196e-06 elapsed=04:09:48 avg_step=1.32s recent_step=1.41s eta=32:36:54 +2026-08-25 08:05:31 | INFO | train rank=0 | step=11330/100000 progress=11.3% loss=0.042603 lr=9.890853e-06 elapsed=04:10:01 avg_step=1.32s recent_step=1.32s eta=32:36:41 +2026-08-25 08:05:42 | INFO | train rank=0 | step=11340/100000 progress=11.3% loss=0.056793 lr=9.890509e-06 elapsed=04:10:12 avg_step=1.32s recent_step=1.16s eta=32:36:15 +2026-08-25 08:05:55 | INFO | train rank=0 | step=11350/100000 progress=11.3% loss=0.165466 lr=9.890164e-06 elapsed=04:10:25 avg_step=1.32s recent_step=1.26s eta=32:35:57 +2026-08-25 08:06:08 | INFO | train rank=0 | step=11360/100000 progress=11.4% loss=0.121277 lr=9.889819e-06 elapsed=04:10:38 avg_step=1.32s recent_step=1.29s eta=32:35:41 +2026-08-25 08:06:23 | INFO | train rank=0 | step=11370/100000 progress=11.4% loss=0.056305 lr=9.889474e-06 elapsed=04:10:53 avg_step=1.32s recent_step=1.49s eta=32:35:40 +2026-08-25 08:06:35 | INFO | train rank=0 | step=11380/100000 progress=11.4% loss=0.063416 lr=9.889128e-06 elapsed=04:11:05 avg_step=1.32s recent_step=1.19s eta=32:35:17 +2026-08-25 08:06:46 | INFO | train rank=0 | step=11390/100000 progress=11.4% loss=0.070557 lr=9.888781e-06 elapsed=04:11:16 avg_step=1.32s recent_step=1.17s eta=32:34:51 +2026-08-25 08:07:03 | INFO | train rank=0 | step=11400/100000 progress=11.4% loss=0.070892 lr=9.888434e-06 elapsed=04:11:33 avg_step=1.32s recent_step=1.67s eta=32:35:05 +2026-08-25 08:07:15 | INFO | train rank=0 | step=11410/100000 progress=11.4% loss=0.150757 lr=9.888087e-06 elapsed=04:11:45 avg_step=1.32s recent_step=1.16s eta=32:34:39 +2026-08-25 08:07:30 | INFO | train rank=0 | step=11420/100000 progress=11.4% loss=0.070984 lr=9.887739e-06 elapsed=04:12:00 avg_step=1.32s recent_step=1.54s eta=32:34:43 +2026-08-25 08:07:48 | INFO | train rank=0 | step=11430/100000 progress=11.4% loss=0.073456 lr=9.887390e-06 elapsed=04:12:18 avg_step=1.32s recent_step=1.78s eta=32:35:05 +2026-08-25 08:07:59 | INFO | train rank=0 | step=11440/100000 progress=11.4% loss=0.065186 lr=9.887041e-06 elapsed=04:12:29 avg_step=1.32s recent_step=1.14s eta=32:34:38 +2026-08-25 08:08:11 | INFO | train rank=0 | step=11450/100000 progress=11.5% loss=0.063232 lr=9.886691e-06 elapsed=04:12:41 avg_step=1.32s recent_step=1.17s eta=32:34:13 +2026-08-25 08:08:25 | INFO | train rank=0 | step=11460/100000 progress=11.5% loss=0.150330 lr=9.886341e-06 elapsed=04:12:55 avg_step=1.32s recent_step=1.41s eta=32:34:06 +2026-08-25 08:08:38 | INFO | train rank=0 | step=11470/100000 progress=11.5% loss=0.071991 lr=9.885990e-06 elapsed=04:13:08 avg_step=1.32s recent_step=1.31s eta=32:33:52 +2026-08-25 08:08:54 | INFO | train rank=0 | step=11480/100000 progress=11.5% loss=0.100433 lr=9.885638e-06 elapsed=04:13:24 avg_step=1.32s recent_step=1.55s eta=32:33:56 +2026-08-25 08:09:08 | INFO | train rank=0 | step=11490/100000 progress=11.5% loss=0.078247 lr=9.885287e-06 elapsed=04:13:38 avg_step=1.32s recent_step=1.39s eta=32:33:48 +2026-08-25 08:09:23 | INFO | train rank=0 | step=11500/100000 progress=11.5% loss=0.058655 lr=9.884934e-06 elapsed=04:13:53 avg_step=1.32s recent_step=1.55s eta=32:33:52 +2026-08-25 08:09:35 | INFO | train rank=0 | step=11510/100000 progress=11.5% loss=0.052200 lr=9.884581e-06 elapsed=04:14:05 avg_step=1.32s recent_step=1.15s eta=32:33:26 +2026-08-25 08:09:53 | INFO | train rank=0 | step=11520/100000 progress=11.5% loss=0.065338 lr=9.884228e-06 elapsed=04:14:23 avg_step=1.32s recent_step=1.82s eta=32:33:50 +2026-08-25 08:10:05 | INFO | train rank=0 | step=11530/100000 progress=11.5% loss=0.051880 lr=9.883874e-06 elapsed=04:14:35 avg_step=1.32s recent_step=1.19s eta=32:33:27 +2026-08-25 08:10:16 | INFO | train rank=0 | step=11540/100000 progress=11.5% loss=0.070831 lr=9.883519e-06 elapsed=04:14:46 avg_step=1.32s recent_step=1.14s eta=32:33:00 +2026-08-25 08:10:28 | INFO | train rank=0 | step=11550/100000 progress=11.6% loss=0.092957 lr=9.883164e-06 elapsed=04:14:58 avg_step=1.32s recent_step=1.18s eta=32:32:35 +2026-08-25 08:10:43 | INFO | train rank=0 | step=11560/100000 progress=11.6% loss=0.074738 lr=9.882808e-06 elapsed=04:15:13 avg_step=1.32s recent_step=1.50s eta=32:32:36 +2026-08-25 08:10:54 | INFO | train rank=0 | step=11570/100000 progress=11.6% loss=0.054962 lr=9.882452e-06 elapsed=04:15:24 avg_step=1.32s recent_step=1.12s eta=32:32:07 +2026-08-25 08:11:09 | INFO | train rank=0 | step=11580/100000 progress=11.6% loss=0.079285 lr=9.882096e-06 elapsed=04:15:39 avg_step=1.32s recent_step=1.51s eta=32:32:08 +2026-08-25 08:11:21 | INFO | train rank=0 | step=11590/100000 progress=11.6% loss=0.049393 lr=9.881738e-06 elapsed=04:15:51 avg_step=1.32s recent_step=1.17s eta=32:31:43 +2026-08-25 08:11:36 | INFO | train rank=0 | step=11600/100000 progress=11.6% loss=0.077393 lr=9.881381e-06 elapsed=04:16:06 avg_step=1.32s recent_step=1.49s eta=32:31:42 +2026-08-25 08:11:48 | INFO | train rank=0 | step=11610/100000 progress=11.6% loss=0.077423 lr=9.881022e-06 elapsed=04:16:18 avg_step=1.32s recent_step=1.18s eta=32:31:18 +2026-08-25 08:11:59 | INFO | train rank=0 | step=11620/100000 progress=11.6% loss=0.065186 lr=9.880663e-06 elapsed=04:16:29 avg_step=1.32s recent_step=1.16s eta=32:30:52 +2026-08-25 08:12:11 | INFO | train rank=0 | step=11630/100000 progress=11.6% loss=0.041428 lr=9.880304e-06 elapsed=04:16:41 avg_step=1.32s recent_step=1.20s eta=32:30:29 +2026-08-25 08:12:24 | INFO | train rank=0 | step=11640/100000 progress=11.6% loss=0.050827 lr=9.879944e-06 elapsed=04:16:54 avg_step=1.32s recent_step=1.29s eta=32:30:13 +2026-08-25 08:12:40 | INFO | train rank=0 | step=11650/100000 progress=11.7% loss=0.059860 lr=9.879584e-06 elapsed=04:17:09 avg_step=1.32s recent_step=1.53s eta=32:30:15 +2026-08-25 08:12:53 | INFO | train rank=0 | step=11660/100000 progress=11.7% loss=0.060989 lr=9.879223e-06 elapsed=04:17:23 avg_step=1.32s recent_step=1.35s eta=32:30:04 +2026-08-25 08:13:06 | INFO | train rank=0 | step=11670/100000 progress=11.7% loss=0.124573 lr=9.878861e-06 elapsed=04:17:35 avg_step=1.32s recent_step=1.25s eta=32:29:45 +2026-08-25 08:13:20 | INFO | train rank=0 | step=11680/100000 progress=11.7% loss=0.041275 lr=9.878499e-06 elapsed=04:17:50 avg_step=1.32s recent_step=1.44s eta=32:29:40 +2026-08-25 08:13:32 | INFO | train rank=0 | step=11690/100000 progress=11.7% loss=0.055634 lr=9.878137e-06 elapsed=04:18:02 avg_step=1.32s recent_step=1.22s eta=32:29:19 +2026-08-25 08:13:47 | INFO | train rank=0 | step=11700/100000 progress=11.7% loss=0.061462 lr=9.877774e-06 elapsed=04:18:17 avg_step=1.32s recent_step=1.45s eta=32:29:16 +2026-08-25 08:14:00 | INFO | train rank=0 | step=11710/100000 progress=11.7% loss=0.053345 lr=9.877410e-06 elapsed=04:18:30 avg_step=1.32s recent_step=1.36s eta=32:29:05 +2026-08-25 08:14:14 | INFO | train rank=0 | step=11720/100000 progress=11.7% loss=0.095398 lr=9.877046e-06 elapsed=04:18:44 avg_step=1.32s recent_step=1.37s eta=32:28:55 +2026-08-25 08:14:27 | INFO | train rank=0 | step=11730/100000 progress=11.7% loss=0.102356 lr=9.876681e-06 elapsed=04:18:57 avg_step=1.32s recent_step=1.31s eta=32:28:41 +2026-08-25 08:14:39 | INFO | train rank=0 | step=11740/100000 progress=11.7% loss=0.136536 lr=9.876316e-06 elapsed=04:19:09 avg_step=1.32s recent_step=1.22s eta=32:28:20 +2026-08-25 08:14:54 | INFO | train rank=0 | step=11750/100000 progress=11.8% loss=0.092194 lr=9.875950e-06 elapsed=04:19:24 avg_step=1.32s recent_step=1.47s eta=32:28:18 +2026-08-25 08:15:05 | INFO | train rank=0 | step=11760/100000 progress=11.8% loss=0.072815 lr=9.875584e-06 elapsed=04:19:35 avg_step=1.32s recent_step=1.15s eta=32:27:52 +2026-08-25 08:15:19 | INFO | train rank=0 | step=11770/100000 progress=11.8% loss=0.069702 lr=9.875217e-06 elapsed=04:19:49 avg_step=1.32s recent_step=1.38s eta=32:27:42 +2026-08-25 08:15:35 | INFO | train rank=0 | step=11780/100000 progress=11.8% loss=0.085846 lr=9.874850e-06 elapsed=04:20:05 avg_step=1.32s recent_step=1.57s eta=32:27:47 +2026-08-25 08:15:47 | INFO | train rank=0 | step=11790/100000 progress=11.8% loss=0.080048 lr=9.874482e-06 elapsed=04:20:17 avg_step=1.32s recent_step=1.22s eta=32:27:26 +2026-08-25 08:15:59 | INFO | train rank=0 | step=11800/100000 progress=11.8% loss=0.084656 lr=9.874113e-06 elapsed=04:20:29 avg_step=1.32s recent_step=1.18s eta=32:27:02 +2026-08-25 08:16:14 | INFO | train rank=0 | step=11810/100000 progress=11.8% loss=0.036804 lr=9.873744e-06 elapsed=04:20:44 avg_step=1.32s recent_step=1.55s eta=32:27:06 +2026-08-25 08:16:27 | INFO | train rank=0 | step=11820/100000 progress=11.8% loss=0.067612 lr=9.873375e-06 elapsed=04:20:57 avg_step=1.32s recent_step=1.23s eta=32:26:45 +2026-08-25 08:16:39 | INFO | train rank=0 | step=11830/100000 progress=11.8% loss=0.143463 lr=9.873005e-06 elapsed=04:21:08 avg_step=1.32s recent_step=1.18s eta=32:26:21 +2026-08-25 08:16:55 | INFO | train rank=0 | step=11840/100000 progress=11.8% loss=0.097961 lr=9.872634e-06 elapsed=04:21:24 avg_step=1.32s recent_step=1.60s eta=32:26:28 +2026-08-25 08:17:06 | INFO | train rank=0 | step=11850/100000 progress=11.8% loss=0.075378 lr=9.872263e-06 elapsed=04:21:36 avg_step=1.32s recent_step=1.16s eta=32:26:03 +2026-08-25 08:17:18 | INFO | train rank=0 | step=11860/100000 progress=11.9% loss=0.046150 lr=9.871892e-06 elapsed=04:21:48 avg_step=1.32s recent_step=1.18s eta=32:25:39 +2026-08-25 08:17:32 | INFO | train rank=0 | step=11870/100000 progress=11.9% loss=0.084381 lr=9.871519e-06 elapsed=04:22:01 avg_step=1.32s recent_step=1.36s eta=32:25:29 +2026-08-25 08:17:43 | INFO | train rank=0 | step=11880/100000 progress=11.9% loss=0.049301 lr=9.871147e-06 elapsed=04:22:13 avg_step=1.32s recent_step=1.17s eta=32:25:04 +2026-08-25 08:17:55 | INFO | train rank=0 | step=11890/100000 progress=11.9% loss=0.060883 lr=9.870774e-06 elapsed=04:22:25 avg_step=1.32s recent_step=1.16s eta=32:24:39 +2026-08-25 08:18:11 | INFO | train rank=0 | step=11900/100000 progress=11.9% loss=0.115723 lr=9.870400e-06 elapsed=04:22:41 avg_step=1.32s recent_step=1.59s eta=32:24:45 +2026-08-25 08:18:26 | INFO | train rank=0 | step=11910/100000 progress=11.9% loss=0.052292 lr=9.870026e-06 elapsed=04:22:56 avg_step=1.32s recent_step=1.52s eta=32:24:46 +2026-08-25 08:18:38 | INFO | train rank=0 | step=11920/100000 progress=11.9% loss=0.084412 lr=9.869651e-06 elapsed=04:23:08 avg_step=1.32s recent_step=1.20s eta=32:24:24 +2026-08-25 08:18:50 | INFO | train rank=0 | step=11930/100000 progress=11.9% loss=0.063828 lr=9.869275e-06 elapsed=04:23:20 avg_step=1.32s recent_step=1.18s eta=32:24:00 +2026-08-25 08:19:04 | INFO | train rank=0 | step=11940/100000 progress=11.9% loss=0.084656 lr=9.868899e-06 elapsed=04:23:34 avg_step=1.32s recent_step=1.44s eta=32:23:56 +2026-08-25 08:19:18 | INFO | train rank=0 | step=11950/100000 progress=11.9% loss=0.038429 lr=9.868523e-06 elapsed=04:23:48 avg_step=1.32s recent_step=1.36s eta=32:23:45 +2026-08-25 08:19:30 | INFO | train rank=0 | step=11960/100000 progress=12.0% loss=0.092987 lr=9.868146e-06 elapsed=04:23:59 avg_step=1.32s recent_step=1.18s eta=32:23:21 +2026-08-25 08:19:41 | INFO | train rank=0 | step=11970/100000 progress=12.0% loss=0.036041 lr=9.867769e-06 elapsed=04:24:11 avg_step=1.32s recent_step=1.16s eta=32:22:55 +2026-08-25 08:19:58 | INFO | train rank=0 | step=11980/100000 progress=12.0% loss=0.063721 lr=9.867391e-06 elapsed=04:24:28 avg_step=1.32s recent_step=1.67s eta=32:23:07 +2026-08-25 08:20:09 | INFO | train rank=0 | step=11990/100000 progress=12.0% loss=0.075531 lr=9.867012e-06 elapsed=04:24:39 avg_step=1.32s recent_step=1.16s eta=32:22:42 +2026-08-25 08:20:23 | INFO | train rank=0 | step=12000/100000 progress=12.0% loss=0.059021 lr=9.866633e-06 elapsed=04:24:53 avg_step=1.32s recent_step=1.34s eta=32:22:30 +2026-08-25 08:20:35 | INFO | train rank=0 | step=12010/100000 progress=12.0% loss=0.057617 lr=9.866253e-06 elapsed=04:25:05 avg_step=1.32s recent_step=1.22s eta=32:22:09 +2026-08-25 08:20:53 | INFO | train rank=0 | step=12020/100000 progress=12.0% loss=0.058533 lr=9.865873e-06 elapsed=04:25:23 avg_step=1.32s recent_step=1.83s eta=32:22:33 +2026-08-25 08:21:05 | INFO | train rank=0 | step=12030/100000 progress=12.0% loss=0.164856 lr=9.865493e-06 elapsed=04:25:35 avg_step=1.32s recent_step=1.19s eta=32:22:10 +2026-08-25 08:21:17 | INFO | train rank=0 | step=12040/100000 progress=12.0% loss=0.128967 lr=9.865111e-06 elapsed=04:25:47 avg_step=1.32s recent_step=1.18s eta=32:21:46 +2026-08-25 08:21:29 | INFO | train rank=0 | step=12050/100000 progress=12.0% loss=0.055450 lr=9.864730e-06 elapsed=04:25:59 avg_step=1.32s recent_step=1.17s eta=32:21:22 +2026-08-25 08:21:41 | INFO | train rank=0 | step=12060/100000 progress=12.1% loss=0.076996 lr=9.864347e-06 elapsed=04:26:11 avg_step=1.32s recent_step=1.24s eta=32:21:03 +2026-08-25 08:21:55 | INFO | train rank=0 | step=12070/100000 progress=12.1% loss=0.080109 lr=9.863965e-06 elapsed=04:26:25 avg_step=1.32s recent_step=1.35s eta=32:20:51 +2026-08-25 08:22:09 | INFO | train rank=0 | step=12080/100000 progress=12.1% loss=0.109802 lr=9.863581e-06 elapsed=04:26:39 avg_step=1.32s recent_step=1.47s eta=32:20:49 +2026-08-25 08:22:21 | INFO | train rank=0 | step=12090/100000 progress=12.1% loss=0.048126 lr=9.863197e-06 elapsed=04:26:51 avg_step=1.32s recent_step=1.19s eta=32:20:26 +2026-08-25 08:22:33 | INFO | train rank=0 | step=12100/100000 progress=12.1% loss=0.052185 lr=9.862813e-06 elapsed=04:27:03 avg_step=1.32s recent_step=1.21s eta=32:20:04 +2026-08-25 08:22:47 | INFO | train rank=0 | step=12110/100000 progress=12.1% loss=0.100403 lr=9.862428e-06 elapsed=04:27:17 avg_step=1.32s recent_step=1.39s eta=32:19:56 +2026-08-25 08:23:04 | INFO | train rank=0 | step=12120/100000 progress=12.1% loss=0.071686 lr=9.862043e-06 elapsed=04:27:34 avg_step=1.32s recent_step=1.70s eta=32:20:10 +2026-08-25 08:23:16 | INFO | train rank=0 | step=12130/100000 progress=12.1% loss=0.054504 lr=9.861657e-06 elapsed=04:27:46 avg_step=1.32s recent_step=1.17s eta=32:19:45 +2026-08-25 08:23:30 | INFO | train rank=0 | step=12140/100000 progress=12.1% loss=0.068298 lr=9.861270e-06 elapsed=04:28:00 avg_step=1.32s recent_step=1.36s eta=32:19:34 +2026-08-25 08:23:44 | INFO | train rank=0 | step=12150/100000 progress=12.2% loss=0.054962 lr=9.860883e-06 elapsed=04:28:14 avg_step=1.32s recent_step=1.45s eta=32:19:30 +2026-08-25 08:23:56 | INFO | train rank=0 | step=12160/100000 progress=12.2% loss=0.060898 lr=9.860495e-06 elapsed=04:28:26 avg_step=1.32s recent_step=1.19s eta=32:19:07 +2026-08-25 08:24:07 | INFO | train rank=0 | step=12170/100000 progress=12.2% loss=0.079376 lr=9.860107e-06 elapsed=04:28:37 avg_step=1.32s recent_step=1.14s eta=32:18:41 +2026-08-25 08:24:19 | INFO | train rank=0 | step=12180/100000 progress=12.2% loss=0.086670 lr=9.859719e-06 elapsed=04:28:49 avg_step=1.32s recent_step=1.19s eta=32:18:18 +2026-08-25 08:24:38 | INFO | train rank=0 | step=12190/100000 progress=12.2% loss=0.067673 lr=9.859329e-06 elapsed=04:29:08 avg_step=1.32s recent_step=1.83s eta=32:18:42 +2026-08-25 08:24:51 | INFO | train rank=0 | step=12200/100000 progress=12.2% loss=0.078918 lr=9.858940e-06 elapsed=04:29:21 avg_step=1.32s recent_step=1.29s eta=32:18:26 +2026-08-25 08:25:05 | INFO | train rank=0 | step=12210/100000 progress=12.2% loss=0.039261 lr=9.858549e-06 elapsed=04:29:35 avg_step=1.32s recent_step=1.40s eta=32:18:18 +2026-08-25 08:25:17 | INFO | train rank=0 | step=12220/100000 progress=12.2% loss=0.071655 lr=9.858159e-06 elapsed=04:29:46 avg_step=1.32s recent_step=1.20s eta=32:17:56 +2026-08-25 08:25:29 | INFO | train rank=0 | step=12230/100000 progress=12.2% loss=0.077332 lr=9.857767e-06 elapsed=04:29:59 avg_step=1.32s recent_step=1.21s eta=32:17:34 +2026-08-25 08:25:43 | INFO | train rank=0 | step=12240/100000 progress=12.2% loss=0.072540 lr=9.857376e-06 elapsed=04:30:13 avg_step=1.32s recent_step=1.47s eta=32:17:31 +2026-08-25 08:25:59 | INFO | train rank=0 | step=12250/100000 progress=12.2% loss=0.142334 lr=9.856983e-06 elapsed=04:30:29 avg_step=1.32s recent_step=1.55s eta=32:17:34 +2026-08-25 08:26:11 | INFO | train rank=0 | step=12260/100000 progress=12.3% loss=0.061554 lr=9.856590e-06 elapsed=04:30:41 avg_step=1.32s recent_step=1.20s eta=32:17:12 +2026-08-25 08:26:25 | INFO | train rank=0 | step=12270/100000 progress=12.3% loss=0.078125 lr=9.856197e-06 elapsed=04:30:55 avg_step=1.32s recent_step=1.39s eta=32:17:03 +2026-08-25 08:26:38 | INFO | train rank=0 | step=12280/100000 progress=12.3% loss=0.061737 lr=9.855803e-06 elapsed=04:31:08 avg_step=1.32s recent_step=1.36s eta=32:16:53 +2026-08-25 08:26:50 | INFO | train rank=0 | step=12290/100000 progress=12.3% loss=0.071198 lr=9.855408e-06 elapsed=04:31:20 avg_step=1.32s recent_step=1.14s eta=32:16:26 +2026-08-25 08:27:04 | INFO | train rank=0 | step=12300/100000 progress=12.3% loss=0.065552 lr=9.855013e-06 elapsed=04:31:34 avg_step=1.32s recent_step=1.46s eta=32:16:23 +2026-08-25 08:27:19 | INFO | train rank=0 | step=12310/100000 progress=12.3% loss=0.038712 lr=9.854618e-06 elapsed=04:31:49 avg_step=1.32s recent_step=1.51s eta=32:16:22 +2026-08-25 08:27:31 | INFO | train rank=0 | step=12320/100000 progress=12.3% loss=0.085327 lr=9.854222e-06 elapsed=04:32:01 avg_step=1.32s recent_step=1.17s eta=32:15:58 +2026-08-25 08:27:43 | INFO | train rank=0 | step=12330/100000 progress=12.3% loss=0.101868 lr=9.853825e-06 elapsed=04:32:13 avg_step=1.32s recent_step=1.19s eta=32:15:35 +2026-08-25 08:27:58 | INFO | train rank=0 | step=12340/100000 progress=12.3% loss=0.068909 lr=9.853428e-06 elapsed=04:32:28 avg_step=1.32s recent_step=1.46s eta=32:15:32 +2026-08-25 08:28:14 | INFO | train rank=0 | step=12350/100000 progress=12.3% loss=0.041359 lr=9.853030e-06 elapsed=04:32:44 avg_step=1.33s recent_step=1.62s eta=32:15:40 +2026-08-25 08:28:26 | INFO | train rank=0 | step=12360/100000 progress=12.4% loss=0.047470 lr=9.852632e-06 elapsed=04:32:56 avg_step=1.32s recent_step=1.20s eta=32:15:18 +2026-08-25 08:28:40 | INFO | train rank=0 | step=12370/100000 progress=12.4% loss=0.143616 lr=9.852233e-06 elapsed=04:33:10 avg_step=1.32s recent_step=1.39s eta=32:15:09 +2026-08-25 08:28:54 | INFO | train rank=0 | step=12380/100000 progress=12.4% loss=0.125732 lr=9.851834e-06 elapsed=04:33:24 avg_step=1.33s recent_step=1.46s eta=32:15:05 +2026-08-25 08:29:07 | INFO | train rank=0 | step=12390/100000 progress=12.4% loss=0.068451 lr=9.851434e-06 elapsed=04:33:37 avg_step=1.33s recent_step=1.29s eta=32:14:50 +2026-08-25 08:29:19 | INFO | train rank=0 | step=12400/100000 progress=12.4% loss=0.062653 lr=9.851034e-06 elapsed=04:33:49 avg_step=1.32s recent_step=1.18s eta=32:14:26 +2026-08-25 08:29:31 | INFO | train rank=0 | step=12410/100000 progress=12.4% loss=0.076294 lr=9.850633e-06 elapsed=04:34:01 avg_step=1.32s recent_step=1.18s eta=32:14:03 +2026-08-25 08:29:45 | INFO | train rank=0 | step=12420/100000 progress=12.4% loss=0.054443 lr=9.850232e-06 elapsed=04:34:15 avg_step=1.32s recent_step=1.39s eta=32:13:54 +2026-08-25 08:30:00 | INFO | train rank=0 | step=12430/100000 progress=12.4% loss=0.061310 lr=9.849830e-06 elapsed=04:34:30 avg_step=1.33s recent_step=1.49s eta=32:13:53 +2026-08-25 08:30:12 | INFO | train rank=0 | step=12440/100000 progress=12.4% loss=0.037643 lr=9.849427e-06 elapsed=04:34:42 avg_step=1.32s recent_step=1.24s eta=32:13:33 +2026-08-25 08:30:24 | INFO | train rank=0 | step=12450/100000 progress=12.4% loss=0.131042 lr=9.849024e-06 elapsed=04:34:54 avg_step=1.32s recent_step=1.17s eta=32:13:09 +2026-08-25 08:30:38 | INFO | train rank=0 | step=12460/100000 progress=12.5% loss=0.045135 lr=9.848621e-06 elapsed=04:35:08 avg_step=1.32s recent_step=1.39s eta=32:13:00 +2026-08-25 08:30:52 | INFO | train rank=0 | step=12470/100000 progress=12.5% loss=0.038055 lr=9.848217e-06 elapsed=04:35:22 avg_step=1.32s recent_step=1.45s eta=32:12:56 +2026-08-25 08:31:03 | INFO | train rank=0 | step=12480/100000 progress=12.5% loss=0.079407 lr=9.847812e-06 elapsed=04:35:33 avg_step=1.32s recent_step=1.09s eta=32:12:26 +2026-08-25 08:31:15 | INFO | train rank=0 | step=12490/100000 progress=12.5% loss=0.077942 lr=9.847407e-06 elapsed=04:35:45 avg_step=1.32s recent_step=1.18s eta=32:12:02 +2026-08-25 08:31:27 | INFO | train rank=0 | step=12500/100000 progress=12.5% loss=0.066986 lr=9.847001e-06 elapsed=04:35:57 avg_step=1.32s recent_step=1.19s eta=32:11:40 +2026-08-25 08:31:40 | INFO | train rank=0 | step=12510/100000 progress=12.5% loss=0.088287 lr=9.846595e-06 elapsed=04:36:10 avg_step=1.32s recent_step=1.33s eta=32:11:27 +2026-08-25 08:31:52 | INFO | train rank=0 | step=12520/100000 progress=12.5% loss=0.046097 lr=9.846188e-06 elapsed=04:36:22 avg_step=1.32s recent_step=1.16s eta=32:11:02 +2026-08-25 08:32:03 | INFO | train rank=0 | step=12530/100000 progress=12.5% loss=0.073212 lr=9.845781e-06 elapsed=04:36:33 avg_step=1.32s recent_step=1.16s eta=32:10:37 +2026-08-25 08:32:16 | INFO | train rank=0 | step=12540/100000 progress=12.5% loss=0.059982 lr=9.845373e-06 elapsed=04:36:46 avg_step=1.32s recent_step=1.25s eta=32:10:19 +2026-08-25 08:32:32 | INFO | train rank=0 | step=12550/100000 progress=12.6% loss=0.072372 lr=9.844965e-06 elapsed=04:37:02 avg_step=1.32s recent_step=1.67s eta=32:10:29 +2026-08-25 08:32:44 | INFO | train rank=0 | step=12560/100000 progress=12.6% loss=0.067780 lr=9.844556e-06 elapsed=04:37:14 avg_step=1.32s recent_step=1.13s eta=32:10:02 +2026-08-25 08:32:56 | INFO | train rank=0 | step=12570/100000 progress=12.6% loss=0.056137 lr=9.844147e-06 elapsed=04:37:25 avg_step=1.32s recent_step=1.18s eta=32:09:39 +2026-08-25 08:33:08 | INFO | train rank=0 | step=12580/100000 progress=12.6% loss=0.083862 lr=9.843737e-06 elapsed=04:37:38 avg_step=1.32s recent_step=1.27s eta=32:09:23 +2026-08-25 08:33:21 | INFO | train rank=0 | step=12590/100000 progress=12.6% loss=0.051926 lr=9.843327e-06 elapsed=04:37:51 avg_step=1.32s recent_step=1.24s eta=32:09:03 +2026-08-25 08:33:33 | INFO | train rank=0 | step=12600/100000 progress=12.6% loss=0.047073 lr=9.842916e-06 elapsed=04:38:03 avg_step=1.32s recent_step=1.25s eta=32:08:45 +2026-08-25 08:33:49 | INFO | train rank=0 | step=12610/100000 progress=12.6% loss=0.079163 lr=9.842504e-06 elapsed=04:38:19 avg_step=1.32s recent_step=1.56s eta=32:08:48 +2026-08-25 08:34:01 | INFO | train rank=0 | step=12620/100000 progress=12.6% loss=0.085968 lr=9.842092e-06 elapsed=04:38:31 avg_step=1.32s recent_step=1.21s eta=32:08:26 +2026-08-25 08:34:18 | INFO | train rank=0 | step=12630/100000 progress=12.6% loss=0.065491 lr=9.841680e-06 elapsed=04:38:48 avg_step=1.32s recent_step=1.73s eta=32:08:41 +2026-08-25 08:34:29 | INFO | train rank=0 | step=12640/100000 progress=12.6% loss=0.096283 lr=9.841267e-06 elapsed=04:38:59 avg_step=1.32s recent_step=1.14s eta=32:08:15 +2026-08-25 08:34:41 | INFO | train rank=0 | step=12650/100000 progress=12.7% loss=0.062820 lr=9.840853e-06 elapsed=04:39:11 avg_step=1.32s recent_step=1.20s eta=32:07:53 +2026-08-25 08:34:55 | INFO | train rank=0 | step=12660/100000 progress=12.7% loss=0.065002 lr=9.840439e-06 elapsed=04:39:25 avg_step=1.32s recent_step=1.36s eta=32:07:42 +2026-08-25 08:35:12 | INFO | train rank=0 | step=12670/100000 progress=12.7% loss=0.063416 lr=9.840024e-06 elapsed=04:39:41 avg_step=1.32s recent_step=1.65s eta=32:07:51 +2026-08-25 08:35:24 | INFO | train rank=0 | step=12680/100000 progress=12.7% loss=0.057480 lr=9.839609e-06 elapsed=04:39:53 avg_step=1.32s recent_step=1.20s eta=32:07:30 +2026-08-25 08:35:35 | INFO | train rank=0 | step=12690/100000 progress=12.7% loss=0.130554 lr=9.839194e-06 elapsed=04:40:05 avg_step=1.32s recent_step=1.15s eta=32:07:05 +2026-08-25 08:35:51 | INFO | train rank=0 | step=12700/100000 progress=12.7% loss=0.071259 lr=9.838777e-06 elapsed=04:40:21 avg_step=1.32s recent_step=1.56s eta=32:07:07 +2026-08-25 08:36:06 | INFO | train rank=0 | step=12710/100000 progress=12.7% loss=0.056854 lr=9.838361e-06 elapsed=04:40:36 avg_step=1.32s recent_step=1.55s eta=32:07:09 +2026-08-25 08:36:18 | INFO | train rank=0 | step=12720/100000 progress=12.7% loss=0.057831 lr=9.837943e-06 elapsed=04:40:48 avg_step=1.32s recent_step=1.16s eta=32:06:45 +2026-08-25 08:36:31 | INFO | train rank=0 | step=12730/100000 progress=12.7% loss=0.062927 lr=9.837525e-06 elapsed=04:41:01 avg_step=1.32s recent_step=1.35s eta=32:06:34 +2026-08-25 08:36:46 | INFO | train rank=0 | step=12740/100000 progress=12.7% loss=0.066132 lr=9.837107e-06 elapsed=04:41:15 avg_step=1.32s recent_step=1.44s eta=32:06:28 +2026-08-25 08:37:01 | INFO | train rank=0 | step=12750/100000 progress=12.8% loss=0.045700 lr=9.836688e-06 elapsed=04:41:31 avg_step=1.32s recent_step=1.55s eta=32:06:30 +2026-08-25 08:37:15 | INFO | train rank=0 | step=12760/100000 progress=12.8% loss=0.094574 lr=9.836269e-06 elapsed=04:41:45 avg_step=1.32s recent_step=1.38s eta=32:06:21 +2026-08-25 08:37:29 | INFO | train rank=0 | step=12770/100000 progress=12.8% loss=0.046509 lr=9.835849e-06 elapsed=04:41:58 avg_step=1.32s recent_step=1.37s eta=32:06:11 +2026-08-25 08:37:40 | INFO | train rank=0 | step=12780/100000 progress=12.8% loss=0.057922 lr=9.835428e-06 elapsed=04:42:10 avg_step=1.32s recent_step=1.17s eta=32:05:47 +2026-08-25 08:37:56 | INFO | train rank=0 | step=12790/100000 progress=12.8% loss=0.043060 lr=9.835007e-06 elapsed=04:42:26 avg_step=1.32s recent_step=1.58s eta=32:05:51 +2026-08-25 08:38:08 | INFO | train rank=0 | step=12800/100000 progress=12.8% loss=0.053711 lr=9.834586e-06 elapsed=04:42:38 avg_step=1.32s recent_step=1.16s eta=32:05:26 +2026-08-25 08:38:21 | INFO | train rank=0 | step=12810/100000 progress=12.8% loss=0.068451 lr=9.834164e-06 elapsed=04:42:51 avg_step=1.32s recent_step=1.37s eta=32:05:16 +2026-08-25 08:38:33 | INFO | train rank=0 | step=12820/100000 progress=12.8% loss=0.045166 lr=9.833741e-06 elapsed=04:43:03 avg_step=1.32s recent_step=1.13s eta=32:04:50 +2026-08-25 08:38:44 | INFO | train rank=0 | step=12830/100000 progress=12.8% loss=0.062500 lr=9.833318e-06 elapsed=04:43:14 avg_step=1.32s recent_step=1.17s eta=32:04:26 +2026-08-25 08:38:56 | INFO | train rank=0 | step=12840/100000 progress=12.8% loss=0.092224 lr=9.832895e-06 elapsed=04:43:26 avg_step=1.32s recent_step=1.16s eta=32:04:01 +2026-08-25 08:39:11 | INFO | train rank=0 | step=12850/100000 progress=12.8% loss=0.075195 lr=9.832470e-06 elapsed=04:43:41 avg_step=1.32s recent_step=1.54s eta=32:04:03 +2026-08-25 08:39:27 | INFO | train rank=0 | step=12860/100000 progress=12.9% loss=0.074249 lr=9.832046e-06 elapsed=04:43:57 avg_step=1.32s recent_step=1.54s eta=32:04:04 +2026-08-25 08:39:38 | INFO | train rank=0 | step=12870/100000 progress=12.9% loss=0.097839 lr=9.831620e-06 elapsed=04:44:08 avg_step=1.32s recent_step=1.15s eta=32:03:39 +2026-08-25 08:39:54 | INFO | train rank=0 | step=12880/100000 progress=12.9% loss=0.048119 lr=9.831195e-06 elapsed=04:44:23 avg_step=1.32s recent_step=1.53s eta=32:03:40 +2026-08-25 08:40:07 | INFO | train rank=0 | step=12890/100000 progress=12.9% loss=0.057907 lr=9.830768e-06 elapsed=04:44:37 avg_step=1.32s recent_step=1.35s eta=32:03:29 +2026-08-25 08:40:22 | INFO | train rank=0 | step=12900/100000 progress=12.9% loss=0.082489 lr=9.830342e-06 elapsed=04:44:52 avg_step=1.33s recent_step=1.51s eta=32:03:28 +2026-08-25 08:40:35 | INFO | train rank=0 | step=12910/100000 progress=12.9% loss=0.064148 lr=9.829914e-06 elapsed=04:45:05 avg_step=1.32s recent_step=1.29s eta=32:03:13 +2026-08-25 08:40:50 | INFO | train rank=0 | step=12920/100000 progress=12.9% loss=0.209351 lr=9.829486e-06 elapsed=04:45:20 avg_step=1.33s recent_step=1.51s eta=32:03:11 +2026-08-25 08:41:02 | INFO | train rank=0 | step=12930/100000 progress=12.9% loss=0.047607 lr=9.829058e-06 elapsed=04:45:32 avg_step=1.33s recent_step=1.23s eta=32:02:52 +2026-08-25 08:41:18 | INFO | train rank=0 | step=12940/100000 progress=12.9% loss=0.071747 lr=9.828629e-06 elapsed=04:45:48 avg_step=1.33s recent_step=1.59s eta=32:02:56 +2026-08-25 08:41:30 | INFO | train rank=0 | step=12950/100000 progress=13.0% loss=0.058090 lr=9.828200e-06 elapsed=04:46:00 avg_step=1.33s recent_step=1.15s eta=32:02:31 +2026-08-25 08:41:42 | INFO | train rank=0 | step=12960/100000 progress=13.0% loss=0.048615 lr=9.827770e-06 elapsed=04:46:12 avg_step=1.33s recent_step=1.20s eta=32:02:10 +2026-08-25 08:41:54 | INFO | train rank=0 | step=12970/100000 progress=13.0% loss=0.056870 lr=9.827339e-06 elapsed=04:46:24 avg_step=1.32s recent_step=1.19s eta=32:01:47 +2026-08-25 08:42:05 | INFO | train rank=0 | step=12980/100000 progress=13.0% loss=0.087799 lr=9.826908e-06 elapsed=04:46:35 avg_step=1.32s recent_step=1.15s eta=32:01:22 +2026-08-25 08:42:18 | INFO | train rank=0 | step=12990/100000 progress=13.0% loss=0.057053 lr=9.826477e-06 elapsed=04:46:47 avg_step=1.32s recent_step=1.22s eta=32:01:02 +2026-08-25 08:42:29 | INFO | train rank=0 | step=13000/100000 progress=13.0% loss=0.067780 lr=9.826045e-06 elapsed=04:46:59 avg_step=1.32s recent_step=1.17s eta=32:00:38 +2026-08-25 08:42:44 | INFO | train rank=0 | step=13010/100000 progress=13.0% loss=0.059418 lr=9.825612e-06 elapsed=04:47:13 avg_step=1.32s recent_step=1.44s eta=32:00:33 +2026-08-25 08:42:56 | INFO | train rank=0 | step=13020/100000 progress=13.0% loss=0.069061 lr=9.825179e-06 elapsed=04:47:26 avg_step=1.32s recent_step=1.23s eta=32:00:13 +2026-08-25 08:43:13 | INFO | train rank=0 | step=13030/100000 progress=13.0% loss=0.070435 lr=9.824745e-06 elapsed=04:47:42 avg_step=1.32s recent_step=1.66s eta=32:00:23 +2026-08-25 08:43:26 | INFO | train rank=0 | step=13040/100000 progress=13.0% loss=0.141052 lr=9.824311e-06 elapsed=04:47:56 avg_step=1.32s recent_step=1.36s eta=32:00:12 +2026-08-25 08:43:38 | INFO | train rank=0 | step=13050/100000 progress=13.1% loss=0.108398 lr=9.823876e-06 elapsed=04:48:08 avg_step=1.32s recent_step=1.22s eta=31:59:52 +2026-08-25 08:43:52 | INFO | train rank=0 | step=13060/100000 progress=13.1% loss=0.097534 lr=9.823441e-06 elapsed=04:48:22 avg_step=1.32s recent_step=1.40s eta=31:59:43 +2026-08-25 08:44:05 | INFO | train rank=0 | step=13070/100000 progress=13.1% loss=0.076477 lr=9.823005e-06 elapsed=04:48:35 avg_step=1.32s recent_step=1.28s eta=31:59:27 +2026-08-25 08:44:20 | INFO | train rank=0 | step=13080/100000 progress=13.1% loss=0.056824 lr=9.822569e-06 elapsed=04:48:50 avg_step=1.32s recent_step=1.50s eta=31:59:25 +2026-08-25 08:44:32 | INFO | train rank=0 | step=13090/100000 progress=13.1% loss=0.046753 lr=9.822132e-06 elapsed=04:49:01 avg_step=1.32s recent_step=1.15s eta=31:59:00 +2026-08-25 08:44:44 | INFO | train rank=0 | step=13100/100000 progress=13.1% loss=0.062332 lr=9.821695e-06 elapsed=04:49:14 avg_step=1.32s recent_step=1.22s eta=31:58:39 +2026-08-25 08:44:55 | INFO | train rank=0 | step=13110/100000 progress=13.1% loss=0.072296 lr=9.821257e-06 elapsed=04:49:25 avg_step=1.32s recent_step=1.11s eta=31:58:12 +2026-08-25 08:45:07 | INFO | train rank=0 | step=13120/100000 progress=13.1% loss=0.076202 lr=9.820818e-06 elapsed=04:49:36 avg_step=1.32s recent_step=1.17s eta=31:57:49 +2026-08-25 08:45:19 | INFO | train rank=0 | step=13130/100000 progress=13.1% loss=0.117126 lr=9.820379e-06 elapsed=04:49:49 avg_step=1.32s recent_step=1.26s eta=31:57:31 +2026-08-25 08:45:30 | INFO | train rank=0 | step=13140/100000 progress=13.1% loss=0.100769 lr=9.819940e-06 elapsed=04:50:00 avg_step=1.32s recent_step=1.13s eta=31:57:05 +2026-08-25 08:45:42 | INFO | train rank=0 | step=13150/100000 progress=13.2% loss=0.062408 lr=9.819500e-06 elapsed=04:50:12 avg_step=1.32s recent_step=1.21s eta=31:56:44 +2026-08-25 08:45:54 | INFO | train rank=0 | step=13160/100000 progress=13.2% loss=0.081116 lr=9.819059e-06 elapsed=04:50:24 avg_step=1.32s recent_step=1.18s eta=31:56:22 +2026-08-25 08:46:06 | INFO | train rank=0 | step=13170/100000 progress=13.2% loss=0.075348 lr=9.818618e-06 elapsed=04:50:35 avg_step=1.32s recent_step=1.13s eta=31:55:55 +2026-08-25 08:46:17 | INFO | train rank=0 | step=13180/100000 progress=13.2% loss=0.059418 lr=9.818177e-06 elapsed=04:50:47 avg_step=1.32s recent_step=1.17s eta=31:55:32 +2026-08-25 08:46:29 | INFO | train rank=0 | step=13190/100000 progress=13.2% loss=0.080170 lr=9.817735e-06 elapsed=04:50:58 avg_step=1.32s recent_step=1.12s eta=31:55:05 +2026-08-25 08:46:42 | INFO | train rank=0 | step=13200/100000 progress=13.2% loss=0.079407 lr=9.817292e-06 elapsed=04:51:12 avg_step=1.32s recent_step=1.38s eta=31:54:56 +2026-08-25 08:46:58 | INFO | train rank=0 | step=13210/100000 progress=13.2% loss=0.106567 lr=9.816849e-06 elapsed=04:51:28 avg_step=1.32s recent_step=1.58s eta=31:54:59 +2026-08-25 08:47:13 | INFO | train rank=0 | step=13220/100000 progress=13.2% loss=0.076263 lr=9.816405e-06 elapsed=04:51:43 avg_step=1.32s recent_step=1.53s eta=31:54:59 +2026-08-25 08:47:25 | INFO | train rank=0 | step=13230/100000 progress=13.2% loss=0.078827 lr=9.815961e-06 elapsed=04:51:55 avg_step=1.32s recent_step=1.17s eta=31:54:36 +2026-08-25 08:47:39 | INFO | train rank=0 | step=13240/100000 progress=13.2% loss=0.092834 lr=9.815516e-06 elapsed=04:52:09 avg_step=1.32s recent_step=1.39s eta=31:54:27 +2026-08-25 08:47:56 | INFO | train rank=0 | step=13250/100000 progress=13.2% loss=0.074158 lr=9.815071e-06 elapsed=04:52:26 avg_step=1.32s recent_step=1.75s eta=31:54:42 +2026-08-25 08:48:09 | INFO | train rank=0 | step=13260/100000 progress=13.3% loss=0.076050 lr=9.814625e-06 elapsed=04:52:38 avg_step=1.32s recent_step=1.21s eta=31:54:21 +2026-08-25 08:48:22 | INFO | train rank=0 | step=13270/100000 progress=13.3% loss=0.070007 lr=9.814179e-06 elapsed=04:52:52 avg_step=1.32s recent_step=1.31s eta=31:54:07 +2026-08-25 08:48:34 | INFO | train rank=0 | step=13280/100000 progress=13.3% loss=0.072052 lr=9.813732e-06 elapsed=04:53:04 avg_step=1.32s recent_step=1.22s eta=31:53:47 +2026-08-25 08:48:46 | INFO | train rank=0 | step=13290/100000 progress=13.3% loss=0.131958 lr=9.813285e-06 elapsed=04:53:16 avg_step=1.32s recent_step=1.22s eta=31:53:27 +2026-08-25 08:48:58 | INFO | train rank=0 | step=13300/100000 progress=13.3% loss=0.077576 lr=9.812837e-06 elapsed=04:53:28 avg_step=1.32s recent_step=1.22s eta=31:53:07 +2026-08-25 08:49:14 | INFO | train rank=0 | step=13310/100000 progress=13.3% loss=0.053299 lr=9.812388e-06 elapsed=04:53:44 avg_step=1.32s recent_step=1.55s eta=31:53:09 +2026-08-25 08:49:28 | INFO | train rank=0 | step=13320/100000 progress=13.3% loss=0.071777 lr=9.811939e-06 elapsed=04:53:58 avg_step=1.32s recent_step=1.39s eta=31:53:00 +2026-08-25 08:49:41 | INFO | train rank=0 | step=13330/100000 progress=13.3% loss=0.055527 lr=9.811490e-06 elapsed=04:54:11 avg_step=1.32s recent_step=1.33s eta=31:52:47 +2026-08-25 08:49:57 | INFO | train rank=0 | step=13340/100000 progress=13.3% loss=0.104736 lr=9.811040e-06 elapsed=04:54:27 avg_step=1.32s recent_step=1.60s eta=31:52:51 +2026-08-25 08:50:13 | INFO | train rank=0 | step=13350/100000 progress=13.4% loss=0.088562 lr=9.810589e-06 elapsed=04:54:43 avg_step=1.32s recent_step=1.60s eta=31:52:56 +2026-08-25 08:50:25 | INFO | train rank=0 | step=13360/100000 progress=13.4% loss=0.065430 lr=9.810138e-06 elapsed=04:54:55 avg_step=1.32s recent_step=1.18s eta=31:52:33 +2026-08-25 08:50:41 | INFO | train rank=0 | step=13370/100000 progress=13.4% loss=0.086182 lr=9.809687e-06 elapsed=04:55:11 avg_step=1.32s recent_step=1.61s eta=31:52:39 +2026-08-25 08:50:54 | INFO | train rank=0 | step=13380/100000 progress=13.4% loss=0.061874 lr=9.809235e-06 elapsed=04:55:24 avg_step=1.32s recent_step=1.35s eta=31:52:27 +2026-08-25 08:51:10 | INFO | train rank=0 | step=13390/100000 progress=13.4% loss=0.064896 lr=9.808782e-06 elapsed=04:55:39 avg_step=1.32s recent_step=1.52s eta=31:52:26 +2026-08-25 08:51:21 | INFO | train rank=0 | step=13400/100000 progress=13.4% loss=0.063232 lr=9.808329e-06 elapsed=04:55:51 avg_step=1.32s recent_step=1.14s eta=31:52:01 +2026-08-25 08:51:37 | INFO | train rank=0 | step=13410/100000 progress=13.4% loss=0.103943 lr=9.807875e-06 elapsed=04:56:07 avg_step=1.32s recent_step=1.60s eta=31:52:06 +2026-08-25 08:51:49 | INFO | train rank=0 | step=13420/100000 progress=13.4% loss=0.060852 lr=9.807421e-06 elapsed=04:56:19 avg_step=1.32s recent_step=1.17s eta=31:51:43 +2026-08-25 08:52:03 | INFO | train rank=0 | step=13430/100000 progress=13.4% loss=0.081512 lr=9.806966e-06 elapsed=04:56:33 avg_step=1.32s recent_step=1.41s eta=31:51:35 +2026-08-25 08:52:15 | INFO | train rank=0 | step=13440/100000 progress=13.4% loss=0.069122 lr=9.806511e-06 elapsed=04:56:45 avg_step=1.32s recent_step=1.21s eta=31:51:14 +2026-08-25 08:52:30 | INFO | train rank=0 | step=13450/100000 progress=13.5% loss=0.078156 lr=9.806055e-06 elapsed=04:56:59 avg_step=1.32s recent_step=1.46s eta=31:51:10 +2026-08-25 08:52:44 | INFO | train rank=0 | step=13460/100000 progress=13.5% loss=0.060684 lr=9.805599e-06 elapsed=04:57:14 avg_step=1.32s recent_step=1.43s eta=31:51:04 +2026-08-25 08:52:56 | INFO | train rank=0 | step=13470/100000 progress=13.5% loss=0.071320 lr=9.805142e-06 elapsed=04:57:26 avg_step=1.32s recent_step=1.22s eta=31:50:43 +2026-08-25 08:53:09 | INFO | train rank=0 | step=13480/100000 progress=13.5% loss=0.066345 lr=9.804685e-06 elapsed=04:57:39 avg_step=1.32s recent_step=1.26s eta=31:50:26 +2026-08-25 08:53:23 | INFO | train rank=0 | step=13490/100000 progress=13.5% loss=0.063568 lr=9.804227e-06 elapsed=04:57:53 avg_step=1.32s recent_step=1.47s eta=31:50:22 +2026-08-25 08:53:39 | INFO | train rank=0 | step=13500/100000 progress=13.5% loss=0.063782 lr=9.803768e-06 elapsed=04:58:09 avg_step=1.33s recent_step=1.62s eta=31:50:27 +2026-08-25 08:53:52 | INFO | train rank=0 | step=13510/100000 progress=13.5% loss=0.053543 lr=9.803309e-06 elapsed=04:58:22 avg_step=1.33s recent_step=1.22s eta=31:50:07 +2026-08-25 08:54:06 | INFO | train rank=0 | step=13520/100000 progress=13.5% loss=0.076172 lr=9.802850e-06 elapsed=04:58:36 avg_step=1.33s recent_step=1.43s eta=31:50:00 +2026-08-25 08:54:20 | INFO | train rank=0 | step=13530/100000 progress=13.5% loss=0.075745 lr=9.802390e-06 elapsed=04:58:50 avg_step=1.33s recent_step=1.39s eta=31:49:51 +2026-08-25 08:54:32 | INFO | train rank=0 | step=13540/100000 progress=13.5% loss=0.092407 lr=9.801929e-06 elapsed=04:59:02 avg_step=1.33s recent_step=1.19s eta=31:49:29 +2026-08-25 08:54:44 | INFO | train rank=0 | step=13550/100000 progress=13.6% loss=0.078400 lr=9.801468e-06 elapsed=04:59:14 avg_step=1.33s recent_step=1.20s eta=31:49:08 +2026-08-25 08:54:55 | INFO | train rank=0 | step=13560/100000 progress=13.6% loss=0.106384 lr=9.801007e-06 elapsed=04:59:25 avg_step=1.32s recent_step=1.13s eta=31:48:42 +2026-08-25 08:55:07 | INFO | train rank=0 | step=13570/100000 progress=13.6% loss=0.050201 lr=9.800545e-06 elapsed=04:59:37 avg_step=1.32s recent_step=1.19s eta=31:48:21 +2026-08-25 08:55:18 | INFO | train rank=0 | step=13580/100000 progress=13.6% loss=0.059921 lr=9.800082e-06 elapsed=04:59:48 avg_step=1.32s recent_step=1.15s eta=31:47:56 +2026-08-25 08:55:30 | INFO | train rank=0 | step=13590/100000 progress=13.6% loss=0.086884 lr=9.799619e-06 elapsed=05:00:00 avg_step=1.32s recent_step=1.20s eta=31:47:35 +2026-08-25 08:55:42 | INFO | train rank=0 | step=13600/100000 progress=13.6% loss=0.072983 lr=9.799155e-06 elapsed=05:00:12 avg_step=1.32s recent_step=1.19s eta=31:47:14 +2026-08-25 08:56:00 | INFO | train rank=0 | step=13610/100000 progress=13.6% loss=0.087830 lr=9.798691e-06 elapsed=05:00:30 avg_step=1.32s recent_step=1.75s eta=31:47:27 +2026-08-25 08:56:17 | INFO | train rank=0 | step=13620/100000 progress=13.6% loss=0.068512 lr=9.798226e-06 elapsed=05:00:47 avg_step=1.33s recent_step=1.74s eta=31:47:40 +2026-08-25 08:56:29 | INFO | train rank=0 | step=13630/100000 progress=13.6% loss=0.054749 lr=9.797761e-06 elapsed=05:00:59 avg_step=1.32s recent_step=1.18s eta=31:47:17 +2026-08-25 08:56:40 | INFO | train rank=0 | step=13640/100000 progress=13.6% loss=0.057953 lr=9.797295e-06 elapsed=05:01:10 avg_step=1.32s recent_step=1.13s eta=31:46:52 +2026-08-25 08:56:54 | INFO | train rank=0 | step=13650/100000 progress=13.7% loss=0.068024 lr=9.796829e-06 elapsed=05:01:24 avg_step=1.32s recent_step=1.41s eta=31:46:44 +2026-08-25 08:57:10 | INFO | train rank=0 | step=13660/100000 progress=13.7% loss=0.056580 lr=9.796362e-06 elapsed=05:01:40 avg_step=1.33s recent_step=1.58s eta=31:46:47 +2026-08-25 08:57:24 | INFO | train rank=0 | step=13670/100000 progress=13.7% loss=0.065704 lr=9.795895e-06 elapsed=05:01:54 avg_step=1.33s recent_step=1.36s eta=31:46:35 +2026-08-25 08:57:35 | INFO | train rank=0 | step=13680/100000 progress=13.7% loss=0.055130 lr=9.795427e-06 elapsed=05:02:05 avg_step=1.32s recent_step=1.17s eta=31:46:12 +2026-08-25 08:57:47 | INFO | train rank=0 | step=13690/100000 progress=13.7% loss=0.066788 lr=9.794959e-06 elapsed=05:02:17 avg_step=1.32s recent_step=1.14s eta=31:45:47 +2026-08-25 08:58:00 | INFO | train rank=0 | step=13700/100000 progress=13.7% loss=0.086121 lr=9.794490e-06 elapsed=05:02:30 avg_step=1.32s recent_step=1.34s eta=31:45:35 +2026-08-25 08:58:12 | INFO | train rank=0 | step=13710/100000 progress=13.7% loss=0.047195 lr=9.794020e-06 elapsed=05:02:42 avg_step=1.32s recent_step=1.18s eta=31:45:12 +2026-08-25 08:58:24 | INFO | train rank=0 | step=13720/100000 progress=13.7% loss=0.061188 lr=9.793550e-06 elapsed=05:02:53 avg_step=1.32s recent_step=1.17s eta=31:44:49 +2026-08-25 08:58:36 | INFO | train rank=0 | step=13730/100000 progress=13.7% loss=0.058762 lr=9.793080e-06 elapsed=05:03:06 avg_step=1.32s recent_step=1.22s eta=31:44:29 +2026-08-25 08:58:53 | INFO | train rank=0 | step=13740/100000 progress=13.7% loss=0.046265 lr=9.792609e-06 elapsed=05:03:23 avg_step=1.32s recent_step=1.76s eta=31:44:43 +2026-08-25 08:59:09 | INFO | train rank=0 | step=13750/100000 progress=13.8% loss=0.065247 lr=9.792137e-06 elapsed=05:03:39 avg_step=1.33s recent_step=1.57s eta=31:44:45 +2026-08-25 08:59:22 | INFO | train rank=0 | step=13760/100000 progress=13.8% loss=0.062286 lr=9.791665e-06 elapsed=05:03:52 avg_step=1.33s recent_step=1.30s eta=31:44:30 +2026-08-25 08:59:34 | INFO | train rank=0 | step=13770/100000 progress=13.8% loss=0.084229 lr=9.791193e-06 elapsed=05:04:04 avg_step=1.32s recent_step=1.17s eta=31:44:07 +2026-08-25 08:59:48 | INFO | train rank=0 | step=13780/100000 progress=13.8% loss=0.056870 lr=9.790720e-06 elapsed=05:04:18 avg_step=1.33s recent_step=1.47s eta=31:44:03 +2026-08-25 09:00:00 | INFO | train rank=0 | step=13790/100000 progress=13.8% loss=0.044052 lr=9.790246e-06 elapsed=05:04:30 avg_step=1.32s recent_step=1.18s eta=31:43:41 +2026-08-25 09:00:16 | INFO | train rank=0 | step=13800/100000 progress=13.8% loss=0.078186 lr=9.789772e-06 elapsed=05:04:46 avg_step=1.33s recent_step=1.54s eta=31:43:41 +2026-08-25 09:00:30 | INFO | train rank=0 | step=13810/100000 progress=13.8% loss=0.046982 lr=9.789297e-06 elapsed=05:05:00 avg_step=1.33s recent_step=1.41s eta=31:43:33 +2026-08-25 09:00:44 | INFO | train rank=0 | step=13820/100000 progress=13.8% loss=0.086914 lr=9.788822e-06 elapsed=05:05:14 avg_step=1.33s recent_step=1.44s eta=31:43:27 +2026-08-25 09:00:56 | INFO | train rank=0 | step=13830/100000 progress=13.8% loss=0.069855 lr=9.788346e-06 elapsed=05:05:26 avg_step=1.33s recent_step=1.17s eta=31:43:04 +2026-08-25 09:01:08 | INFO | train rank=0 | step=13840/100000 progress=13.8% loss=0.113892 lr=9.787870e-06 elapsed=05:05:37 avg_step=1.32s recent_step=1.17s eta=31:42:41 +2026-08-25 09:01:22 | INFO | train rank=0 | step=13850/100000 progress=13.9% loss=0.071075 lr=9.787393e-06 elapsed=05:05:52 avg_step=1.33s recent_step=1.45s eta=31:42:36 +2026-08-25 09:01:34 | INFO | train rank=0 | step=13860/100000 progress=13.9% loss=0.095764 lr=9.786916e-06 elapsed=05:06:04 avg_step=1.33s recent_step=1.21s eta=31:42:16 +2026-08-25 09:01:49 | INFO | train rank=0 | step=13870/100000 progress=13.9% loss=0.158569 lr=9.786438e-06 elapsed=05:06:19 avg_step=1.33s recent_step=1.47s eta=31:42:12 +2026-08-25 09:02:04 | INFO | train rank=0 | step=13880/100000 progress=13.9% loss=0.109558 lr=9.785960e-06 elapsed=05:06:34 avg_step=1.33s recent_step=1.50s eta=31:42:09 +2026-08-25 09:02:15 | INFO | train rank=0 | step=13890/100000 progress=13.9% loss=0.065338 lr=9.785481e-06 elapsed=05:06:45 avg_step=1.33s recent_step=1.13s eta=31:41:44 +2026-08-25 09:02:27 | INFO | train rank=0 | step=13900/100000 progress=13.9% loss=0.060242 lr=9.785002e-06 elapsed=05:06:57 avg_step=1.32s recent_step=1.17s eta=31:41:21 +2026-08-25 09:02:40 | INFO | train rank=0 | step=13910/100000 progress=13.9% loss=0.055023 lr=9.784522e-06 elapsed=05:07:10 avg_step=1.33s recent_step=1.35s eta=31:41:10 +2026-08-25 09:02:52 | INFO | train rank=0 | step=13920/100000 progress=13.9% loss=0.067841 lr=9.784041e-06 elapsed=05:07:22 avg_step=1.32s recent_step=1.13s eta=31:40:44 +2026-08-25 09:03:10 | INFO | train rank=0 | step=13930/100000 progress=13.9% loss=0.098816 lr=9.783560e-06 elapsed=05:07:40 avg_step=1.33s recent_step=1.84s eta=31:41:03 +2026-08-25 09:03:22 | INFO | train rank=0 | step=13940/100000 progress=13.9% loss=0.156372 lr=9.783079e-06 elapsed=05:07:52 avg_step=1.33s recent_step=1.22s eta=31:40:43 +2026-08-25 09:03:36 | INFO | train rank=0 | step=13950/100000 progress=14.0% loss=0.046921 lr=9.782597e-06 elapsed=05:08:05 avg_step=1.33s recent_step=1.33s eta=31:40:30 +2026-08-25 09:03:51 | INFO | train rank=0 | step=13960/100000 progress=14.0% loss=0.054901 lr=9.782114e-06 elapsed=05:08:21 avg_step=1.33s recent_step=1.57s eta=31:40:31 +2026-08-25 09:04:03 | INFO | train rank=0 | step=13970/100000 progress=14.0% loss=0.046066 lr=9.781631e-06 elapsed=05:08:33 avg_step=1.33s recent_step=1.17s eta=31:40:09 +2026-08-25 09:04:21 | INFO | train rank=0 | step=13980/100000 progress=14.0% loss=0.063400 lr=9.781148e-06 elapsed=05:08:51 avg_step=1.33s recent_step=1.79s eta=31:40:24 +2026-08-25 09:04:33 | INFO | train rank=0 | step=13990/100000 progress=14.0% loss=0.081116 lr=9.780664e-06 elapsed=05:09:02 avg_step=1.33s recent_step=1.16s eta=31:40:01 +2026-08-25 09:04:48 | INFO | train rank=0 | step=14000/100000 progress=14.0% loss=0.068085 lr=9.780179e-06 elapsed=05:09:18 avg_step=1.33s recent_step=1.59s eta=31:40:03 +2026-08-25 09:05:01 | INFO | train rank=0 | step=14010/100000 progress=14.0% loss=0.040421 lr=9.779694e-06 elapsed=05:09:31 avg_step=1.33s recent_step=1.26s eta=31:39:46 +2026-08-25 09:05:13 | INFO | train rank=0 | step=14020/100000 progress=14.0% loss=0.055359 lr=9.779208e-06 elapsed=05:09:43 avg_step=1.33s recent_step=1.23s eta=31:39:27 +2026-08-25 09:05:27 | INFO | train rank=0 | step=14030/100000 progress=14.0% loss=0.061752 lr=9.778722e-06 elapsed=05:09:57 avg_step=1.33s recent_step=1.38s eta=31:39:17 +2026-08-25 09:05:39 | INFO | train rank=0 | step=14040/100000 progress=14.0% loss=0.065353 lr=9.778235e-06 elapsed=05:10:09 avg_step=1.33s recent_step=1.17s eta=31:38:54 +2026-08-25 09:05:51 | INFO | train rank=0 | step=14050/100000 progress=14.1% loss=0.052704 lr=9.777748e-06 elapsed=05:10:21 avg_step=1.33s recent_step=1.22s eta=31:38:35 +2026-08-25 09:06:09 | INFO | train rank=0 | step=14060/100000 progress=14.1% loss=0.067352 lr=9.777260e-06 elapsed=05:10:39 avg_step=1.33s recent_step=1.76s eta=31:38:48 +2026-08-25 09:06:21 | INFO | train rank=0 | step=14070/100000 progress=14.1% loss=0.071075 lr=9.776772e-06 elapsed=05:10:50 avg_step=1.33s recent_step=1.20s eta=31:38:27 +2026-08-25 09:06:32 | INFO | train rank=0 | step=14080/100000 progress=14.1% loss=0.047760 lr=9.776283e-06 elapsed=05:11:02 avg_step=1.33s recent_step=1.16s eta=31:38:04 +2026-08-25 09:06:44 | INFO | train rank=0 | step=14090/100000 progress=14.1% loss=0.063446 lr=9.775794e-06 elapsed=05:11:14 avg_step=1.33s recent_step=1.19s eta=31:37:42 +2026-08-25 09:06:56 | INFO | train rank=0 | step=14100/100000 progress=14.1% loss=0.065750 lr=9.775304e-06 elapsed=05:11:26 avg_step=1.33s recent_step=1.24s eta=31:37:24 +2026-08-25 09:07:09 | INFO | train rank=0 | step=14110/100000 progress=14.1% loss=0.058716 lr=9.774814e-06 elapsed=05:11:39 avg_step=1.33s recent_step=1.23s eta=31:37:04 +2026-08-25 09:07:21 | INFO | train rank=0 | step=14120/100000 progress=14.1% loss=0.055389 lr=9.774323e-06 elapsed=05:11:50 avg_step=1.33s recent_step=1.18s eta=31:36:42 +2026-08-25 09:07:32 | INFO | train rank=0 | step=14130/100000 progress=14.1% loss=0.056854 lr=9.773831e-06 elapsed=05:12:02 avg_step=1.33s recent_step=1.17s eta=31:36:20 +2026-08-25 09:07:47 | INFO | train rank=0 | step=14140/100000 progress=14.1% loss=0.056549 lr=9.773339e-06 elapsed=05:12:17 avg_step=1.33s recent_step=1.51s eta=31:36:18 +2026-08-25 09:07:59 | INFO | train rank=0 | step=14150/100000 progress=14.1% loss=0.075226 lr=9.772847e-06 elapsed=05:12:29 avg_step=1.33s recent_step=1.15s eta=31:35:54 +2026-08-25 09:08:13 | INFO | train rank=0 | step=14160/100000 progress=14.2% loss=0.044235 lr=9.772354e-06 elapsed=05:12:43 avg_step=1.33s recent_step=1.42s eta=31:35:47 +2026-08-25 09:08:26 | INFO | train rank=0 | step=14170/100000 progress=14.2% loss=0.049026 lr=9.771860e-06 elapsed=05:12:56 avg_step=1.33s recent_step=1.26s eta=31:35:29 +2026-08-25 09:08:40 | INFO | train rank=0 | step=14180/100000 progress=14.2% loss=0.086487 lr=9.771366e-06 elapsed=05:13:10 avg_step=1.33s recent_step=1.41s eta=31:35:21 +2026-08-25 09:08:51 | INFO | train rank=0 | step=14190/100000 progress=14.2% loss=0.060242 lr=9.770872e-06 elapsed=05:13:21 avg_step=1.32s recent_step=1.16s eta=31:34:58 +2026-08-25 09:09:08 | INFO | train rank=0 | step=14200/100000 progress=14.2% loss=0.082123 lr=9.770377e-06 elapsed=05:13:38 avg_step=1.33s recent_step=1.68s eta=31:35:06 +2026-08-25 09:09:20 | INFO | train rank=0 | step=14210/100000 progress=14.2% loss=0.058014 lr=9.769881e-06 elapsed=05:13:50 avg_step=1.33s recent_step=1.17s eta=31:34:43 +2026-08-25 09:09:36 | INFO | train rank=0 | step=14220/100000 progress=14.2% loss=0.068695 lr=9.769385e-06 elapsed=05:14:06 avg_step=1.33s recent_step=1.64s eta=31:34:49 +2026-08-25 09:09:48 | INFO | train rank=0 | step=14230/100000 progress=14.2% loss=0.054611 lr=9.768889e-06 elapsed=05:14:18 avg_step=1.33s recent_step=1.14s eta=31:34:25 +2026-08-25 09:10:03 | INFO | train rank=0 | step=14240/100000 progress=14.2% loss=0.055115 lr=9.768391e-06 elapsed=05:14:33 avg_step=1.33s recent_step=1.50s eta=31:34:22 +2026-08-25 09:10:15 | INFO | train rank=0 | step=14250/100000 progress=14.2% loss=0.058319 lr=9.767894e-06 elapsed=05:14:44 avg_step=1.33s recent_step=1.19s eta=31:34:00 +2026-08-25 09:10:28 | INFO | train rank=0 | step=14260/100000 progress=14.3% loss=0.064331 lr=9.767395e-06 elapsed=05:14:58 avg_step=1.33s recent_step=1.37s eta=31:33:50 +2026-08-25 09:10:42 | INFO | train rank=0 | step=14270/100000 progress=14.3% loss=0.066956 lr=9.766897e-06 elapsed=05:15:12 avg_step=1.33s recent_step=1.37s eta=31:33:39 +2026-08-25 09:10:57 | INFO | train rank=0 | step=14280/100000 progress=14.3% loss=0.053436 lr=9.766398e-06 elapsed=05:15:27 avg_step=1.33s recent_step=1.48s eta=31:33:35 +2026-08-25 09:11:09 | INFO | train rank=0 | step=14290/100000 progress=14.3% loss=0.122314 lr=9.765898e-06 elapsed=05:15:38 avg_step=1.33s recent_step=1.18s eta=31:33:13 +2026-08-25 09:11:22 | INFO | train rank=0 | step=14300/100000 progress=14.3% loss=0.061523 lr=9.765398e-06 elapsed=05:15:52 avg_step=1.33s recent_step=1.37s eta=31:33:03 +2026-08-25 09:11:36 | INFO | train rank=0 | step=14310/100000 progress=14.3% loss=0.051666 lr=9.764897e-06 elapsed=05:16:06 avg_step=1.33s recent_step=1.41s eta=31:32:55 +2026-08-25 09:11:50 | INFO | train rank=0 | step=14320/100000 progress=14.3% loss=0.094482 lr=9.764395e-06 elapsed=05:16:20 avg_step=1.33s recent_step=1.40s eta=31:32:46 +2026-08-25 09:12:02 | INFO | train rank=0 | step=14330/100000 progress=14.3% loss=0.055756 lr=9.763894e-06 elapsed=05:16:32 avg_step=1.33s recent_step=1.18s eta=31:32:24 +2026-08-25 09:12:17 | INFO | train rank=0 | step=14340/100000 progress=14.3% loss=0.073685 lr=9.763391e-06 elapsed=05:16:47 avg_step=1.33s recent_step=1.47s eta=31:32:19 +2026-08-25 09:12:31 | INFO | train rank=0 | step=14350/100000 progress=14.3% loss=0.075134 lr=9.762888e-06 elapsed=05:17:01 avg_step=1.33s recent_step=1.40s eta=31:32:10 +2026-08-25 09:12:42 | INFO | train rank=0 | step=14360/100000 progress=14.4% loss=0.065094 lr=9.762385e-06 elapsed=05:17:12 avg_step=1.33s recent_step=1.11s eta=31:31:44 +2026-08-25 09:12:56 | INFO | train rank=0 | step=14370/100000 progress=14.4% loss=0.085510 lr=9.761881e-06 elapsed=05:17:26 avg_step=1.33s recent_step=1.39s eta=31:31:35 +2026-08-25 09:13:11 | INFO | train rank=0 | step=14380/100000 progress=14.4% loss=0.113159 lr=9.761377e-06 elapsed=05:17:41 avg_step=1.33s recent_step=1.48s eta=31:31:31 +2026-08-25 09:13:22 | INFO | train rank=0 | step=14390/100000 progress=14.4% loss=0.077881 lr=9.760872e-06 elapsed=05:17:52 avg_step=1.33s recent_step=1.14s eta=31:31:06 +2026-08-25 09:13:34 | INFO | train rank=0 | step=14400/100000 progress=14.4% loss=0.091217 lr=9.760366e-06 elapsed=05:18:03 avg_step=1.33s recent_step=1.15s eta=31:30:43 +2026-08-25 09:13:45 | INFO | train rank=0 | step=14410/100000 progress=14.4% loss=0.083984 lr=9.759860e-06 elapsed=05:18:15 avg_step=1.33s recent_step=1.19s eta=31:30:22 +2026-08-25 09:13:59 | INFO | train rank=0 | step=14420/100000 progress=14.4% loss=0.067963 lr=9.759354e-06 elapsed=05:18:29 avg_step=1.33s recent_step=1.37s eta=31:30:11 +2026-08-25 09:14:15 | INFO | train rank=0 | step=14430/100000 progress=14.4% loss=0.050644 lr=9.758846e-06 elapsed=05:18:44 avg_step=1.33s recent_step=1.54s eta=31:30:11 +2026-08-25 09:14:29 | INFO | train rank=0 | step=14440/100000 progress=14.4% loss=0.065948 lr=9.758339e-06 elapsed=05:18:58 avg_step=1.33s recent_step=1.39s eta=31:30:02 +2026-08-25 09:14:41 | INFO | train rank=0 | step=14450/100000 progress=14.4% loss=0.087189 lr=9.757831e-06 elapsed=05:19:11 avg_step=1.33s recent_step=1.23s eta=31:29:43 +2026-08-25 09:14:53 | INFO | train rank=0 | step=14460/100000 progress=14.5% loss=0.075500 lr=9.757322e-06 elapsed=05:19:23 avg_step=1.33s recent_step=1.26s eta=31:29:26 +2026-08-25 09:15:08 | INFO | train rank=0 | step=14470/100000 progress=14.5% loss=0.051941 lr=9.756813e-06 elapsed=05:19:38 avg_step=1.33s recent_step=1.48s eta=31:29:21 +2026-08-25 09:15:20 | INFO | train rank=0 | step=14480/100000 progress=14.5% loss=0.064575 lr=9.756303e-06 elapsed=05:19:50 avg_step=1.33s recent_step=1.19s eta=31:29:00 +2026-08-25 09:15:32 | INFO | train rank=0 | step=14490/100000 progress=14.5% loss=0.063751 lr=9.755793e-06 elapsed=05:20:02 avg_step=1.33s recent_step=1.16s eta=31:28:37 +2026-08-25 09:15:46 | INFO | train rank=0 | step=14500/100000 progress=14.5% loss=0.061081 lr=9.755283e-06 elapsed=05:20:16 avg_step=1.33s recent_step=1.48s eta=31:28:33 +2026-08-25 09:15:58 | INFO | train rank=0 | step=14510/100000 progress=14.5% loss=0.094086 lr=9.754771e-06 elapsed=05:20:28 avg_step=1.33s recent_step=1.19s eta=31:28:12 +2026-08-25 09:16:14 | INFO | train rank=0 | step=14520/100000 progress=14.5% loss=0.051132 lr=9.754260e-06 elapsed=05:20:43 avg_step=1.33s recent_step=1.52s eta=31:28:09 +2026-08-25 09:16:26 | INFO | train rank=0 | step=14530/100000 progress=14.5% loss=0.030060 lr=9.753747e-06 elapsed=05:20:56 avg_step=1.33s recent_step=1.23s eta=31:27:51 +2026-08-25 09:16:43 | INFO | train rank=0 | step=14540/100000 progress=14.5% loss=0.063782 lr=9.753235e-06 elapsed=05:21:13 avg_step=1.33s recent_step=1.72s eta=31:28:01 +2026-08-25 09:17:00 | INFO | train rank=0 | step=14550/100000 progress=14.5% loss=0.074951 lr=9.752721e-06 elapsed=05:21:30 avg_step=1.33s recent_step=1.73s eta=31:28:11 +2026-08-25 09:17:12 | INFO | train rank=0 | step=14560/100000 progress=14.6% loss=0.058136 lr=9.752208e-06 elapsed=05:21:42 avg_step=1.33s recent_step=1.20s eta=31:27:51 +2026-08-25 09:17:24 | INFO | train rank=0 | step=14570/100000 progress=14.6% loss=0.054382 lr=9.751693e-06 elapsed=05:21:54 avg_step=1.33s recent_step=1.15s eta=31:27:27 +2026-08-25 09:17:36 | INFO | train rank=0 | step=14580/100000 progress=14.6% loss=0.085205 lr=9.751178e-06 elapsed=05:22:06 avg_step=1.33s recent_step=1.17s eta=31:27:05 +2026-08-25 09:17:47 | INFO | train rank=0 | step=14590/100000 progress=14.6% loss=0.081512 lr=9.750663e-06 elapsed=05:22:17 avg_step=1.33s recent_step=1.17s eta=31:26:43 +2026-08-25 09:17:59 | INFO | train rank=0 | step=14600/100000 progress=14.6% loss=0.079559 lr=9.750147e-06 elapsed=05:22:29 avg_step=1.33s recent_step=1.13s eta=31:26:18 +2026-08-25 09:18:11 | INFO | train rank=0 | step=14610/100000 progress=14.6% loss=0.088196 lr=9.749631e-06 elapsed=05:22:41 avg_step=1.33s recent_step=1.20s eta=31:25:58 +2026-08-25 09:18:22 | INFO | train rank=0 | step=14620/100000 progress=14.6% loss=0.051788 lr=9.749114e-06 elapsed=05:22:52 avg_step=1.33s recent_step=1.15s eta=31:25:34 +2026-08-25 09:18:34 | INFO | train rank=0 | step=14630/100000 progress=14.6% loss=0.068787 lr=9.748596e-06 elapsed=05:23:04 avg_step=1.32s recent_step=1.17s eta=31:25:12 +2026-08-25 09:18:51 | INFO | train rank=0 | step=14640/100000 progress=14.6% loss=0.076324 lr=9.748078e-06 elapsed=05:23:21 avg_step=1.33s recent_step=1.71s eta=31:25:21 +2026-08-25 09:19:03 | INFO | train rank=0 | step=14650/100000 progress=14.6% loss=0.045624 lr=9.747560e-06 elapsed=05:23:33 avg_step=1.33s recent_step=1.21s eta=31:25:01 +2026-08-25 09:19:19 | INFO | train rank=0 | step=14660/100000 progress=14.7% loss=0.054092 lr=9.747041e-06 elapsed=05:23:49 avg_step=1.33s recent_step=1.57s eta=31:25:02 +2026-08-25 09:19:30 | INFO | train rank=0 | step=14670/100000 progress=14.7% loss=0.066711 lr=9.746521e-06 elapsed=05:24:00 avg_step=1.33s recent_step=1.17s eta=31:24:40 +2026-08-25 09:19:46 | INFO | train rank=0 | step=14680/100000 progress=14.7% loss=0.039658 lr=9.746001e-06 elapsed=05:24:16 avg_step=1.33s recent_step=1.59s eta=31:24:42 +2026-08-25 09:19:58 | INFO | train rank=0 | step=14690/100000 progress=14.7% loss=0.090668 lr=9.745481e-06 elapsed=05:24:28 avg_step=1.33s recent_step=1.20s eta=31:24:21 +2026-08-25 09:20:16 | INFO | train rank=0 | step=14700/100000 progress=14.7% loss=0.061279 lr=9.744960e-06 elapsed=05:24:46 avg_step=1.33s recent_step=1.76s eta=31:24:33 +2026-08-25 09:20:28 | INFO | train rank=0 | step=14710/100000 progress=14.7% loss=0.046570 lr=9.744438e-06 elapsed=05:24:58 avg_step=1.33s recent_step=1.19s eta=31:24:12 +2026-08-25 09:20:41 | INFO | train rank=0 | step=14720/100000 progress=14.7% loss=0.056778 lr=9.743916e-06 elapsed=05:25:11 avg_step=1.33s recent_step=1.34s eta=31:24:00 +2026-08-25 09:20:53 | INFO | train rank=0 | step=14730/100000 progress=14.7% loss=0.046371 lr=9.743393e-06 elapsed=05:25:23 avg_step=1.33s recent_step=1.21s eta=31:23:40 +2026-08-25 09:21:06 | INFO | train rank=0 | step=14740/100000 progress=14.7% loss=0.048401 lr=9.742870e-06 elapsed=05:25:36 avg_step=1.33s recent_step=1.31s eta=31:23:26 +2026-08-25 09:21:20 | INFO | train rank=0 | step=14750/100000 progress=14.8% loss=0.065735 lr=9.742346e-06 elapsed=05:25:50 avg_step=1.33s recent_step=1.37s eta=31:23:15 +2026-08-25 09:21:32 | INFO | train rank=0 | step=14760/100000 progress=14.8% loss=0.062439 lr=9.741822e-06 elapsed=05:26:02 avg_step=1.33s recent_step=1.22s eta=31:22:56 +2026-08-25 09:21:45 | INFO | train rank=0 | step=14770/100000 progress=14.8% loss=0.108093 lr=9.741298e-06 elapsed=05:26:15 avg_step=1.33s recent_step=1.23s eta=31:22:37 +2026-08-25 09:21:57 | INFO | train rank=0 | step=14780/100000 progress=14.8% loss=0.062897 lr=9.740772e-06 elapsed=05:26:27 avg_step=1.33s recent_step=1.28s eta=31:22:21 +2026-08-25 09:22:09 | INFO | train rank=0 | step=14790/100000 progress=14.8% loss=0.071228 lr=9.740247e-06 elapsed=05:26:39 avg_step=1.33s recent_step=1.20s eta=31:22:01 +2026-08-25 09:22:22 | INFO | train rank=0 | step=14800/100000 progress=14.8% loss=0.060471 lr=9.739720e-06 elapsed=05:26:52 avg_step=1.33s recent_step=1.24s eta=31:21:42 +2026-08-25 09:22:36 | INFO | train rank=0 | step=14810/100000 progress=14.8% loss=0.038528 lr=9.739194e-06 elapsed=05:27:06 avg_step=1.33s recent_step=1.39s eta=31:21:33 +2026-08-25 09:22:49 | INFO | train rank=0 | step=14820/100000 progress=14.8% loss=0.042770 lr=9.738666e-06 elapsed=05:27:18 avg_step=1.33s recent_step=1.27s eta=31:21:17 +2026-08-25 09:23:00 | INFO | train rank=0 | step=14830/100000 progress=14.8% loss=0.089508 lr=9.738138e-06 elapsed=05:27:30 avg_step=1.33s recent_step=1.18s eta=31:20:55 +2026-08-25 09:23:15 | INFO | train rank=0 | step=14840/100000 progress=14.8% loss=0.145874 lr=9.737610e-06 elapsed=05:27:45 avg_step=1.33s recent_step=1.50s eta=31:20:52 +2026-08-25 09:23:27 | INFO | train rank=0 | step=14850/100000 progress=14.8% loss=0.072327 lr=9.737081e-06 elapsed=05:27:57 avg_step=1.33s recent_step=1.22s eta=31:20:33 +2026-08-25 09:23:46 | INFO | train rank=0 | step=14860/100000 progress=14.9% loss=0.050385 lr=9.736552e-06 elapsed=05:28:16 avg_step=1.33s recent_step=1.82s eta=31:20:48 +2026-08-25 09:23:58 | INFO | train rank=0 | step=14870/100000 progress=14.9% loss=0.043182 lr=9.736022e-06 elapsed=05:28:28 avg_step=1.33s recent_step=1.23s eta=31:20:29 +2026-08-25 09:24:10 | INFO | train rank=0 | step=14880/100000 progress=14.9% loss=0.075470 lr=9.735492e-06 elapsed=05:28:40 avg_step=1.33s recent_step=1.23s eta=31:20:10 +2026-08-25 09:24:22 | INFO | train rank=0 | step=14890/100000 progress=14.9% loss=0.097473 lr=9.734961e-06 elapsed=05:28:52 avg_step=1.33s recent_step=1.21s eta=31:19:50 +2026-08-25 09:24:37 | INFO | train rank=0 | step=14900/100000 progress=14.9% loss=0.056763 lr=9.734429e-06 elapsed=05:29:07 avg_step=1.33s recent_step=1.45s eta=31:19:44 +2026-08-25 09:24:49 | INFO | train rank=0 | step=14910/100000 progress=14.9% loss=0.073090 lr=9.733897e-06 elapsed=05:29:19 avg_step=1.33s recent_step=1.19s eta=31:19:23 +2026-08-25 09:25:01 | INFO | train rank=0 | step=14920/100000 progress=14.9% loss=0.059753 lr=9.733365e-06 elapsed=05:29:31 avg_step=1.33s recent_step=1.20s eta=31:19:03 +2026-08-25 09:25:13 | INFO | train rank=0 | step=14930/100000 progress=14.9% loss=0.118774 lr=9.732832e-06 elapsed=05:29:42 avg_step=1.33s recent_step=1.18s eta=31:18:41 +2026-08-25 09:25:27 | INFO | train rank=0 | step=14940/100000 progress=14.9% loss=0.099792 lr=9.732298e-06 elapsed=05:29:57 avg_step=1.33s recent_step=1.46s eta=31:18:36 +2026-08-25 09:25:39 | INFO | train rank=0 | step=14950/100000 progress=14.9% loss=0.080933 lr=9.731764e-06 elapsed=05:30:09 avg_step=1.33s recent_step=1.23s eta=31:18:17 +2026-08-25 09:25:52 | INFO | train rank=0 | step=14960/100000 progress=15.0% loss=0.090454 lr=9.731230e-06 elapsed=05:30:22 avg_step=1.33s recent_step=1.22s eta=31:17:58 +2026-08-25 09:26:08 | INFO | train rank=0 | step=14970/100000 progress=15.0% loss=0.076080 lr=9.730695e-06 elapsed=05:30:38 avg_step=1.33s recent_step=1.60s eta=31:18:00 +2026-08-25 09:26:20 | INFO | train rank=0 | step=14980/100000 progress=15.0% loss=0.080078 lr=9.730159e-06 elapsed=05:30:50 avg_step=1.33s recent_step=1.27s eta=31:17:44 +2026-08-25 09:26:35 | INFO | train rank=0 | step=14990/100000 progress=15.0% loss=0.055099 lr=9.729623e-06 elapsed=05:31:05 avg_step=1.33s recent_step=1.43s eta=31:17:36 +2026-08-25 09:26:47 | INFO | train rank=0 | step=15000/100000 progress=15.0% loss=0.057190 lr=9.729086e-06 elapsed=05:31:17 avg_step=1.33s recent_step=1.21s eta=31:17:16 +2026-08-25 09:26:59 | INFO | train rank=0 | step=15010/100000 progress=15.0% loss=0.095123 lr=9.728549e-06 elapsed=05:31:29 avg_step=1.33s recent_step=1.27s eta=31:17:00 +2026-08-25 09:27:12 | INFO | train rank=0 | step=15020/100000 progress=15.0% loss=0.039810 lr=9.728011e-06 elapsed=05:31:41 avg_step=1.33s recent_step=1.22s eta=31:16:41 +2026-08-25 09:27:26 | INFO | train rank=0 | step=15030/100000 progress=15.0% loss=0.050781 lr=9.727473e-06 elapsed=05:31:56 avg_step=1.33s recent_step=1.41s eta=31:16:32 +2026-08-25 09:27:37 | INFO | train rank=0 | step=15040/100000 progress=15.0% loss=0.073547 lr=9.726935e-06 elapsed=05:32:07 avg_step=1.32s recent_step=1.18s eta=31:16:10 +2026-08-25 09:27:50 | INFO | train rank=0 | step=15050/100000 progress=15.0% loss=0.066132 lr=9.726395e-06 elapsed=05:32:20 avg_step=1.32s recent_step=1.30s eta=31:15:56 +2026-08-25 09:28:06 | INFO | train rank=0 | step=15060/100000 progress=15.1% loss=0.052002 lr=9.725856e-06 elapsed=05:32:36 avg_step=1.33s recent_step=1.60s eta=31:15:58 +2026-08-25 09:28:18 | INFO | train rank=0 | step=15070/100000 progress=15.1% loss=0.057739 lr=9.725315e-06 elapsed=05:32:48 avg_step=1.33s recent_step=1.19s eta=31:15:37 +2026-08-25 09:28:30 | INFO | train rank=0 | step=15080/100000 progress=15.1% loss=0.076782 lr=9.724775e-06 elapsed=05:33:00 avg_step=1.32s recent_step=1.19s eta=31:15:16 +2026-08-25 09:28:46 | INFO | train rank=0 | step=15090/100000 progress=15.1% loss=0.079132 lr=9.724233e-06 elapsed=05:33:16 avg_step=1.33s recent_step=1.60s eta=31:15:18 +2026-08-25 09:28:58 | INFO | train rank=0 | step=15100/100000 progress=15.1% loss=0.065338 lr=9.723692e-06 elapsed=05:33:28 avg_step=1.33s recent_step=1.21s eta=31:14:59 +2026-08-25 09:29:13 | INFO | train rank=0 | step=15110/100000 progress=15.1% loss=0.087769 lr=9.723149e-06 elapsed=05:33:43 avg_step=1.33s recent_step=1.51s eta=31:14:56 +2026-08-25 09:29:26 | INFO | train rank=0 | step=15120/100000 progress=15.1% loss=0.045227 lr=9.722606e-06 elapsed=05:33:56 avg_step=1.33s recent_step=1.23s eta=31:14:37 +2026-08-25 09:29:41 | INFO | train rank=0 | step=15130/100000 progress=15.1% loss=0.056976 lr=9.722063e-06 elapsed=05:34:11 avg_step=1.33s recent_step=1.51s eta=31:14:34 +2026-08-25 09:29:53 | INFO | train rank=0 | step=15140/100000 progress=15.1% loss=0.121643 lr=9.721519e-06 elapsed=05:34:23 avg_step=1.33s recent_step=1.21s eta=31:14:14 +2026-08-25 09:30:04 | INFO | train rank=0 | step=15150/100000 progress=15.2% loss=0.071548 lr=9.720975e-06 elapsed=05:34:34 avg_step=1.33s recent_step=1.17s eta=31:13:52 +2026-08-25 09:30:17 | INFO | train rank=0 | step=15160/100000 progress=15.2% loss=0.064331 lr=9.720430e-06 elapsed=05:34:46 avg_step=1.32s recent_step=1.20s eta=31:13:32 +2026-08-25 09:30:32 | INFO | train rank=0 | step=15170/100000 progress=15.2% loss=0.067017 lr=9.719885e-06 elapsed=05:35:02 avg_step=1.33s recent_step=1.58s eta=31:13:33 +2026-08-25 09:30:46 | INFO | train rank=0 | step=15180/100000 progress=15.2% loss=0.070099 lr=9.719339e-06 elapsed=05:35:16 avg_step=1.33s recent_step=1.37s eta=31:13:22 +2026-08-25 09:31:00 | INFO | train rank=0 | step=15190/100000 progress=15.2% loss=0.086731 lr=9.718792e-06 elapsed=05:35:30 avg_step=1.33s recent_step=1.42s eta=31:13:14 +2026-08-25 09:31:12 | INFO | train rank=0 | step=15200/100000 progress=15.2% loss=0.069641 lr=9.718245e-06 elapsed=05:35:42 avg_step=1.33s recent_step=1.18s eta=31:12:53 +2026-08-25 09:31:25 | INFO | train rank=0 | step=15210/100000 progress=15.2% loss=0.039070 lr=9.717698e-06 elapsed=05:35:55 avg_step=1.33s recent_step=1.34s eta=31:12:41 +2026-08-25 09:31:38 | INFO | train rank=0 | step=15220/100000 progress=15.2% loss=0.085510 lr=9.717150e-06 elapsed=05:36:08 avg_step=1.33s recent_step=1.25s eta=31:12:23 +2026-08-25 09:31:50 | INFO | train rank=0 | step=15230/100000 progress=15.2% loss=0.053696 lr=9.716601e-06 elapsed=05:36:20 avg_step=1.33s recent_step=1.19s eta=31:12:02 +2026-08-25 09:32:03 | INFO | train rank=0 | step=15240/100000 progress=15.2% loss=0.084412 lr=9.716052e-06 elapsed=05:36:33 avg_step=1.33s recent_step=1.30s eta=31:11:47 +2026-08-25 09:32:16 | INFO | train rank=0 | step=15250/100000 progress=15.2% loss=0.072266 lr=9.715503e-06 elapsed=05:36:46 avg_step=1.33s recent_step=1.32s eta=31:11:34 +2026-08-25 09:32:31 | INFO | train rank=0 | step=15260/100000 progress=15.3% loss=0.100494 lr=9.714953e-06 elapsed=05:37:01 avg_step=1.33s recent_step=1.50s eta=31:11:30 +2026-08-25 09:32:44 | INFO | train rank=0 | step=15270/100000 progress=15.3% loss=0.108856 lr=9.714402e-06 elapsed=05:37:14 avg_step=1.33s recent_step=1.27s eta=31:11:14 +2026-08-25 09:33:02 | INFO | train rank=0 | step=15280/100000 progress=15.3% loss=0.042969 lr=9.713851e-06 elapsed=05:37:32 avg_step=1.33s recent_step=1.86s eta=31:11:30 +2026-08-25 09:33:14 | INFO | train rank=0 | step=15290/100000 progress=15.3% loss=0.047348 lr=9.713299e-06 elapsed=05:37:44 avg_step=1.33s recent_step=1.20s eta=31:11:10 +2026-08-25 09:33:26 | INFO | train rank=0 | step=15300/100000 progress=15.3% loss=0.059021 lr=9.712747e-06 elapsed=05:37:56 avg_step=1.33s recent_step=1.17s eta=31:10:48 +2026-08-25 09:33:37 | INFO | train rank=0 | step=15310/100000 progress=15.3% loss=0.055588 lr=9.712195e-06 elapsed=05:38:07 avg_step=1.33s recent_step=1.13s eta=31:10:24 +2026-08-25 09:33:49 | INFO | train rank=0 | step=15320/100000 progress=15.3% loss=0.062744 lr=9.711642e-06 elapsed=05:38:19 avg_step=1.33s recent_step=1.18s eta=31:10:03 +2026-08-25 09:34:05 | INFO | train rank=0 | step=15330/100000 progress=15.3% loss=0.069595 lr=9.711088e-06 elapsed=05:38:35 avg_step=1.33s recent_step=1.64s eta=31:10:07 +2026-08-25 09:34:17 | INFO | train rank=0 | step=15340/100000 progress=15.3% loss=0.081085 lr=9.710534e-06 elapsed=05:38:46 avg_step=1.33s recent_step=1.12s eta=31:09:42 +2026-08-25 09:34:29 | INFO | train rank=0 | step=15350/100000 progress=15.3% loss=0.073059 lr=9.709979e-06 elapsed=05:38:59 avg_step=1.33s recent_step=1.22s eta=31:09:23 +2026-08-25 09:34:40 | INFO | train rank=0 | step=15360/100000 progress=15.4% loss=0.062714 lr=9.709424e-06 elapsed=05:39:10 avg_step=1.32s recent_step=1.12s eta=31:08:58 +2026-08-25 09:34:53 | INFO | train rank=0 | step=15370/100000 progress=15.4% loss=0.070160 lr=9.708868e-06 elapsed=05:39:23 avg_step=1.32s recent_step=1.28s eta=31:08:43 +2026-08-25 09:35:05 | INFO | train rank=0 | step=15380/100000 progress=15.4% loss=0.086121 lr=9.708312e-06 elapsed=05:39:35 avg_step=1.32s recent_step=1.20s eta=31:08:22 +2026-08-25 09:35:22 | INFO | train rank=0 | step=15390/100000 progress=15.4% loss=0.059219 lr=9.707755e-06 elapsed=05:39:52 avg_step=1.33s recent_step=1.77s eta=31:08:33 +2026-08-25 09:35:35 | INFO | train rank=0 | step=15400/100000 progress=15.4% loss=0.078827 lr=9.707198e-06 elapsed=05:40:05 avg_step=1.33s recent_step=1.30s eta=31:08:19 +2026-08-25 09:35:49 | INFO | train rank=0 | step=15410/100000 progress=15.4% loss=0.101746 lr=9.706640e-06 elapsed=05:40:19 avg_step=1.33s recent_step=1.37s eta=31:08:08 +2026-08-25 09:36:03 | INFO | train rank=0 | step=15420/100000 progress=15.4% loss=0.102783 lr=9.706082e-06 elapsed=05:40:33 avg_step=1.33s recent_step=1.35s eta=31:07:56 +2026-08-25 09:36:18 | INFO | train rank=0 | step=15430/100000 progress=15.4% loss=0.072968 lr=9.705523e-06 elapsed=05:40:48 avg_step=1.33s recent_step=1.51s eta=31:07:53 +2026-08-25 09:36:30 | INFO | train rank=0 | step=15440/100000 progress=15.4% loss=0.061539 lr=9.704964e-06 elapsed=05:41:00 avg_step=1.33s recent_step=1.18s eta=31:07:32 +2026-08-25 09:36:44 | INFO | train rank=0 | step=15450/100000 progress=15.4% loss=0.127625 lr=9.704404e-06 elapsed=05:41:14 avg_step=1.33s recent_step=1.44s eta=31:07:25 +2026-08-25 09:36:59 | INFO | train rank=0 | step=15460/100000 progress=15.5% loss=0.068848 lr=9.703843e-06 elapsed=05:41:29 avg_step=1.33s recent_step=1.48s eta=31:07:21 +2026-08-25 09:37:12 | INFO | train rank=0 | step=15470/100000 progress=15.5% loss=0.054977 lr=9.703283e-06 elapsed=05:41:42 avg_step=1.33s recent_step=1.33s eta=31:07:08 +2026-08-25 09:37:26 | INFO | train rank=0 | step=15480/100000 progress=15.5% loss=0.056427 lr=9.702721e-06 elapsed=05:41:56 avg_step=1.33s recent_step=1.37s eta=31:06:57 +2026-08-25 09:37:44 | INFO | train rank=0 | step=15490/100000 progress=15.5% loss=0.077057 lr=9.702159e-06 elapsed=05:42:14 avg_step=1.33s recent_step=1.79s eta=31:07:09 +2026-08-25 09:37:55 | INFO | train rank=0 | step=15500/100000 progress=15.5% loss=0.071869 lr=9.701597e-06 elapsed=05:42:25 avg_step=1.33s recent_step=1.17s eta=31:06:47 +2026-08-25 09:38:07 | INFO | train rank=0 | step=15510/100000 progress=15.5% loss=0.057739 lr=9.701034e-06 elapsed=05:42:37 avg_step=1.33s recent_step=1.14s eta=31:06:24 +2026-08-25 09:38:22 | INFO | train rank=0 | step=15520/100000 progress=15.5% loss=0.069977 lr=9.700471e-06 elapsed=05:42:52 avg_step=1.33s recent_step=1.49s eta=31:06:20 +2026-08-25 09:38:33 | INFO | train rank=0 | step=15530/100000 progress=15.5% loss=0.055435 lr=9.699907e-06 elapsed=05:43:03 avg_step=1.33s recent_step=1.16s eta=31:05:57 +2026-08-25 09:38:45 | INFO | train rank=0 | step=15540/100000 progress=15.5% loss=0.047958 lr=9.699342e-06 elapsed=05:43:15 avg_step=1.33s recent_step=1.18s eta=31:05:36 +2026-08-25 09:38:57 | INFO | train rank=0 | step=15550/100000 progress=15.6% loss=0.061798 lr=9.698777e-06 elapsed=05:43:26 avg_step=1.33s recent_step=1.14s eta=31:05:13 +2026-08-25 09:39:09 | INFO | train rank=0 | step=15560/100000 progress=15.6% loss=0.060257 lr=9.698212e-06 elapsed=05:43:39 avg_step=1.33s recent_step=1.22s eta=31:04:54 +2026-08-25 09:39:21 | INFO | train rank=0 | step=15570/100000 progress=15.6% loss=0.117432 lr=9.697646e-06 elapsed=05:43:51 avg_step=1.33s recent_step=1.27s eta=31:04:38 +2026-08-25 09:39:33 | INFO | train rank=0 | step=15580/100000 progress=15.6% loss=0.082825 lr=9.697079e-06 elapsed=05:44:03 avg_step=1.33s recent_step=1.18s eta=31:04:17 +2026-08-25 09:39:48 | INFO | train rank=0 | step=15590/100000 progress=15.6% loss=0.050644 lr=9.696512e-06 elapsed=05:44:18 avg_step=1.33s recent_step=1.44s eta=31:04:10 +2026-08-25 09:40:01 | INFO | train rank=0 | step=15600/100000 progress=15.6% loss=0.056030 lr=9.695945e-06 elapsed=05:44:31 avg_step=1.33s recent_step=1.36s eta=31:03:58 +2026-08-25 09:40:13 | INFO | train rank=0 | step=15610/100000 progress=15.6% loss=0.058502 lr=9.695377e-06 elapsed=05:44:43 avg_step=1.33s recent_step=1.22s eta=31:03:39 +2026-08-25 09:40:28 | INFO | train rank=0 | step=15620/100000 progress=15.6% loss=0.057220 lr=9.694808e-06 elapsed=05:44:58 avg_step=1.33s recent_step=1.47s eta=31:03:34 +2026-08-25 09:40:43 | INFO | train rank=0 | step=15630/100000 progress=15.6% loss=0.043121 lr=9.694239e-06 elapsed=05:45:13 avg_step=1.33s recent_step=1.46s eta=31:03:27 +2026-08-25 09:40:55 | INFO | train rank=0 | step=15640/100000 progress=15.6% loss=0.052429 lr=9.693669e-06 elapsed=05:45:25 avg_step=1.33s recent_step=1.25s eta=31:03:10 +2026-08-25 09:41:07 | INFO | train rank=0 | step=15650/100000 progress=15.7% loss=0.071777 lr=9.693099e-06 elapsed=05:45:37 avg_step=1.33s recent_step=1.20s eta=31:02:50 +2026-08-25 09:41:22 | INFO | train rank=0 | step=15660/100000 progress=15.7% loss=0.058594 lr=9.692529e-06 elapsed=05:45:51 avg_step=1.33s recent_step=1.44s eta=31:02:43 +2026-08-25 09:41:34 | INFO | train rank=0 | step=15670/100000 progress=15.7% loss=0.065552 lr=9.691957e-06 elapsed=05:46:04 avg_step=1.33s recent_step=1.22s eta=31:02:25 +2026-08-25 09:41:47 | INFO | train rank=0 | step=15680/100000 progress=15.7% loss=0.069794 lr=9.691386e-06 elapsed=05:46:17 avg_step=1.33s recent_step=1.36s eta=31:02:13 +2026-08-25 09:42:01 | INFO | train rank=0 | step=15690/100000 progress=15.7% loss=0.086487 lr=9.690814e-06 elapsed=05:46:31 avg_step=1.33s recent_step=1.38s eta=31:02:03 +2026-08-25 09:42:14 | INFO | train rank=0 | step=15700/100000 progress=15.7% loss=0.057144 lr=9.690241e-06 elapsed=05:46:44 avg_step=1.33s recent_step=1.27s eta=31:01:47 +2026-08-25 09:42:26 | INFO | train rank=0 | step=15710/100000 progress=15.7% loss=0.068237 lr=9.689668e-06 elapsed=05:46:56 avg_step=1.33s recent_step=1.19s eta=31:01:26 +2026-08-25 09:42:42 | INFO | train rank=0 | step=15720/100000 progress=15.7% loss=0.085754 lr=9.689094e-06 elapsed=05:47:12 avg_step=1.33s recent_step=1.61s eta=31:01:28 +2026-08-25 09:42:53 | INFO | train rank=0 | step=15730/100000 progress=15.7% loss=0.051491 lr=9.688520e-06 elapsed=05:47:23 avg_step=1.33s recent_step=1.11s eta=31:01:03 +2026-08-25 09:43:05 | INFO | train rank=0 | step=15740/100000 progress=15.7% loss=0.052231 lr=9.687945e-06 elapsed=05:47:35 avg_step=1.33s recent_step=1.25s eta=31:00:46 +2026-08-25 09:43:21 | INFO | train rank=0 | step=15750/100000 progress=15.8% loss=0.119080 lr=9.687370e-06 elapsed=05:47:50 avg_step=1.33s recent_step=1.51s eta=31:00:42 +2026-08-25 09:43:34 | INFO | train rank=0 | step=15760/100000 progress=15.8% loss=0.069855 lr=9.686794e-06 elapsed=05:48:04 avg_step=1.33s recent_step=1.37s eta=31:00:32 +2026-08-25 09:43:48 | INFO | train rank=0 | step=15770/100000 progress=15.8% loss=0.054337 lr=9.686218e-06 elapsed=05:48:18 avg_step=1.33s recent_step=1.37s eta=31:00:21 +2026-08-25 09:44:02 | INFO | train rank=0 | step=15780/100000 progress=15.8% loss=0.066772 lr=9.685641e-06 elapsed=05:48:32 avg_step=1.33s recent_step=1.43s eta=31:00:13 +2026-08-25 09:44:17 | INFO | train rank=0 | step=15790/100000 progress=15.8% loss=0.057373 lr=9.685064e-06 elapsed=05:48:47 avg_step=1.33s recent_step=1.49s eta=31:00:09 +2026-08-25 09:44:29 | INFO | train rank=0 | step=15800/100000 progress=15.8% loss=0.040619 lr=9.684486e-06 elapsed=05:48:59 avg_step=1.33s recent_step=1.19s eta=30:59:49 +2026-08-25 09:44:40 | INFO | train rank=0 | step=15810/100000 progress=15.8% loss=0.055817 lr=9.683908e-06 elapsed=05:49:10 avg_step=1.33s recent_step=1.12s eta=30:59:24 +2026-08-25 09:44:53 | INFO | train rank=0 | step=15820/100000 progress=15.8% loss=0.113983 lr=9.683329e-06 elapsed=05:49:23 avg_step=1.33s recent_step=1.24s eta=30:59:06 +2026-08-25 09:45:05 | INFO | train rank=0 | step=15830/100000 progress=15.8% loss=0.045799 lr=9.682749e-06 elapsed=05:49:35 avg_step=1.33s recent_step=1.21s eta=30:58:47 +2026-08-25 09:45:21 | INFO | train rank=0 | step=15840/100000 progress=15.8% loss=0.080200 lr=9.682170e-06 elapsed=05:49:51 avg_step=1.33s recent_step=1.64s eta=30:58:51 +2026-08-25 09:45:33 | INFO | train rank=0 | step=15850/100000 progress=15.8% loss=0.042068 lr=9.681589e-06 elapsed=05:50:03 avg_step=1.33s recent_step=1.16s eta=30:58:29 +2026-08-25 09:45:49 | INFO | train rank=0 | step=15860/100000 progress=15.9% loss=0.054901 lr=9.681008e-06 elapsed=05:50:19 avg_step=1.33s recent_step=1.59s eta=30:58:29 +2026-08-25 09:46:01 | INFO | train rank=0 | step=15870/100000 progress=15.9% loss=0.066895 lr=9.680427e-06 elapsed=05:50:31 avg_step=1.33s recent_step=1.21s eta=30:58:10 +2026-08-25 09:46:13 | INFO | train rank=0 | step=15880/100000 progress=15.9% loss=0.051788 lr=9.679845e-06 elapsed=05:50:43 avg_step=1.33s recent_step=1.20s eta=30:57:50 +2026-08-25 09:46:30 | INFO | train rank=0 | step=15890/100000 progress=15.9% loss=0.059357 lr=9.679263e-06 elapsed=05:51:00 avg_step=1.33s recent_step=1.74s eta=30:57:59 +2026-08-25 09:46:43 | INFO | train rank=0 | step=15900/100000 progress=15.9% loss=0.054413 lr=9.678680e-06 elapsed=05:51:12 avg_step=1.33s recent_step=1.23s eta=30:57:41 +2026-08-25 09:46:54 | INFO | train rank=0 | step=15910/100000 progress=15.9% loss=0.100433 lr=9.678096e-06 elapsed=05:51:24 avg_step=1.33s recent_step=1.17s eta=30:57:20 +2026-08-25 09:47:08 | INFO | train rank=0 | step=15920/100000 progress=15.9% loss=0.109863 lr=9.677512e-06 elapsed=05:51:38 avg_step=1.33s recent_step=1.37s eta=30:57:09 +2026-08-25 09:47:20 | INFO | train rank=0 | step=15930/100000 progress=15.9% loss=0.051224 lr=9.676928e-06 elapsed=05:51:50 avg_step=1.33s recent_step=1.20s eta=30:56:49 +2026-08-25 09:47:36 | INFO | train rank=0 | step=15940/100000 progress=15.9% loss=0.051025 lr=9.676343e-06 elapsed=05:52:06 avg_step=1.33s recent_step=1.61s eta=30:56:51 +2026-08-25 09:47:50 | INFO | train rank=0 | step=15950/100000 progress=16.0% loss=0.053955 lr=9.675757e-06 elapsed=05:52:20 avg_step=1.33s recent_step=1.42s eta=30:56:42 +2026-08-25 09:48:04 | INFO | train rank=0 | step=15960/100000 progress=16.0% loss=0.042862 lr=9.675171e-06 elapsed=05:52:34 avg_step=1.33s recent_step=1.36s eta=30:56:31 +2026-08-25 09:48:17 | INFO | train rank=0 | step=15970/100000 progress=16.0% loss=0.053131 lr=9.674585e-06 elapsed=05:52:47 avg_step=1.33s recent_step=1.34s eta=30:56:18 +2026-08-25 09:48:33 | INFO | train rank=0 | step=15980/100000 progress=16.0% loss=0.086060 lr=9.673998e-06 elapsed=05:53:03 avg_step=1.33s recent_step=1.58s eta=30:56:18 +2026-08-25 09:48:44 | INFO | train rank=0 | step=15990/100000 progress=16.0% loss=0.072052 lr=9.673410e-06 elapsed=05:53:14 avg_step=1.33s recent_step=1.11s eta=30:55:54 +2026-08-25 09:48:56 | INFO | train rank=0 | step=16000/100000 progress=16.0% loss=0.066833 lr=9.672822e-06 elapsed=05:53:26 avg_step=1.33s recent_step=1.20s eta=30:55:34 +2026-08-25 09:49:08 | INFO | train rank=0 | step=16010/100000 progress=16.0% loss=0.041382 lr=9.672234e-06 elapsed=05:53:37 avg_step=1.33s recent_step=1.15s eta=30:55:11 +2026-08-25 09:49:24 | INFO | train rank=0 | step=16020/100000 progress=16.0% loss=0.052956 lr=9.671645e-06 elapsed=05:53:54 avg_step=1.33s recent_step=1.67s eta=30:55:16 +2026-08-25 09:49:37 | INFO | train rank=0 | step=16030/100000 progress=16.0% loss=0.056976 lr=9.671055e-06 elapsed=05:54:06 avg_step=1.33s recent_step=1.23s eta=30:54:57 +2026-08-25 09:49:52 | INFO | train rank=0 | step=16040/100000 progress=16.0% loss=0.082001 lr=9.670465e-06 elapsed=05:54:22 avg_step=1.33s recent_step=1.58s eta=30:54:58 +2026-08-25 09:50:09 | INFO | train rank=0 | step=16050/100000 progress=16.1% loss=0.046921 lr=9.669874e-06 elapsed=05:54:39 avg_step=1.33s recent_step=1.68s eta=30:55:03 +2026-08-25 09:50:23 | INFO | train rank=0 | step=16060/100000 progress=16.1% loss=0.084198 lr=9.669283e-06 elapsed=05:54:53 avg_step=1.33s recent_step=1.43s eta=30:54:55 +2026-08-25 09:50:38 | INFO | train rank=0 | step=16070/100000 progress=16.1% loss=0.115173 lr=9.668692e-06 elapsed=05:55:07 avg_step=1.33s recent_step=1.41s eta=30:54:46 +2026-08-25 09:50:50 | INFO | train rank=0 | step=16080/100000 progress=16.1% loss=0.042465 lr=9.668100e-06 elapsed=05:55:20 avg_step=1.33s recent_step=1.28s eta=30:54:31 +2026-08-25 09:51:05 | INFO | train rank=0 | step=16090/100000 progress=16.1% loss=0.045135 lr=9.667507e-06 elapsed=05:55:35 avg_step=1.33s recent_step=1.48s eta=30:54:25 +2026-08-25 09:51:16 | INFO | train rank=0 | step=16100/100000 progress=16.1% loss=0.078064 lr=9.666914e-06 elapsed=05:55:46 avg_step=1.33s recent_step=1.12s eta=30:54:01 +2026-08-25 09:51:28 | INFO | train rank=0 | step=16110/100000 progress=16.1% loss=0.104095 lr=9.666320e-06 elapsed=05:55:58 avg_step=1.33s recent_step=1.19s eta=30:53:41 +2026-08-25 09:51:39 | INFO | train rank=0 | step=16120/100000 progress=16.1% loss=0.049347 lr=9.665726e-06 elapsed=05:56:09 avg_step=1.33s recent_step=1.09s eta=30:53:16 +2026-08-25 09:51:59 | INFO | train rank=0 | step=16130/100000 progress=16.1% loss=0.047668 lr=9.665131e-06 elapsed=05:56:28 avg_step=1.33s recent_step=1.93s eta=30:53:34 +2026-08-25 09:52:10 | INFO | train rank=0 | step=16140/100000 progress=16.1% loss=0.052979 lr=9.664536e-06 elapsed=05:56:40 avg_step=1.33s recent_step=1.18s eta=30:53:13 +2026-08-25 09:52:21 | INFO | train rank=0 | step=16150/100000 progress=16.2% loss=0.097900 lr=9.663940e-06 elapsed=05:56:51 avg_step=1.33s recent_step=1.10s eta=30:52:49 +2026-08-25 09:52:41 | INFO | train rank=0 | step=16160/100000 progress=16.2% loss=0.064697 lr=9.663344e-06 elapsed=05:57:11 avg_step=1.33s recent_step=1.95s eta=30:53:08 +2026-08-25 09:52:54 | INFO | train rank=0 | step=16170/100000 progress=16.2% loss=0.048584 lr=9.662748e-06 elapsed=05:57:24 avg_step=1.33s recent_step=1.32s eta=30:52:54 +2026-08-25 09:53:05 | INFO | train rank=0 | step=16180/100000 progress=16.2% loss=0.091614 lr=9.662150e-06 elapsed=05:57:35 avg_step=1.33s recent_step=1.13s eta=30:52:30 +2026-08-25 09:53:19 | INFO | train rank=0 | step=16190/100000 progress=16.2% loss=0.068878 lr=9.661553e-06 elapsed=05:57:49 avg_step=1.33s recent_step=1.38s eta=30:52:20 +2026-08-25 09:53:34 | INFO | train rank=0 | step=16200/100000 progress=16.2% loss=0.066315 lr=9.660954e-06 elapsed=05:58:04 avg_step=1.33s recent_step=1.45s eta=30:52:13 +2026-08-25 09:53:47 | INFO | train rank=0 | step=16210/100000 progress=16.2% loss=0.083923 lr=9.660356e-06 elapsed=05:58:16 avg_step=1.33s recent_step=1.29s eta=30:51:58 +2026-08-25 09:53:58 | INFO | train rank=0 | step=16220/100000 progress=16.2% loss=0.034683 lr=9.659756e-06 elapsed=05:58:28 avg_step=1.33s recent_step=1.14s eta=30:51:35 +2026-08-25 09:54:12 | INFO | train rank=0 | step=16230/100000 progress=16.2% loss=0.057648 lr=9.659157e-06 elapsed=05:58:42 avg_step=1.33s recent_step=1.43s eta=30:51:27 +2026-08-25 09:54:25 | INFO | train rank=0 | step=16240/100000 progress=16.2% loss=0.052521 lr=9.658556e-06 elapsed=05:58:54 avg_step=1.33s recent_step=1.23s eta=30:51:09 +2026-08-25 09:54:39 | INFO | train rank=0 | step=16250/100000 progress=16.2% loss=0.067993 lr=9.657955e-06 elapsed=05:59:09 avg_step=1.33s recent_step=1.43s eta=30:51:01 +2026-08-25 09:54:50 | INFO | train rank=0 | step=16260/100000 progress=16.3% loss=0.051682 lr=9.657354e-06 elapsed=05:59:20 avg_step=1.33s recent_step=1.15s eta=30:50:39 +2026-08-25 09:55:02 | INFO | train rank=0 | step=16270/100000 progress=16.3% loss=0.060791 lr=9.656752e-06 elapsed=05:59:32 avg_step=1.33s recent_step=1.17s eta=30:50:18 +2026-08-25 09:55:13 | INFO | train rank=0 | step=16280/100000 progress=16.3% loss=0.069016 lr=9.656150e-06 elapsed=05:59:43 avg_step=1.33s recent_step=1.11s eta=30:49:53 +2026-08-25 09:55:28 | INFO | train rank=0 | step=16290/100000 progress=16.3% loss=0.065964 lr=9.655547e-06 elapsed=05:59:57 avg_step=1.33s recent_step=1.43s eta=30:49:46 +2026-08-25 09:55:39 | INFO | train rank=0 | step=16300/100000 progress=16.3% loss=0.038330 lr=9.654944e-06 elapsed=06:00:09 avg_step=1.33s recent_step=1.16s eta=30:49:24 +2026-08-25 09:55:51 | INFO | train rank=0 | step=16310/100000 progress=16.3% loss=0.100952 lr=9.654340e-06 elapsed=06:00:20 avg_step=1.33s recent_step=1.15s eta=30:49:01 +2026-08-25 09:56:06 | INFO | train rank=0 | step=16320/100000 progress=16.3% loss=0.040314 lr=9.653736e-06 elapsed=06:00:36 avg_step=1.33s recent_step=1.56s eta=30:49:00 +2026-08-25 09:56:23 | INFO | train rank=0 | step=16330/100000 progress=16.3% loss=0.059540 lr=9.653131e-06 elapsed=06:00:53 avg_step=1.33s recent_step=1.66s eta=30:49:04 +2026-08-25 09:56:36 | INFO | train rank=0 | step=16340/100000 progress=16.3% loss=0.107788 lr=9.652525e-06 elapsed=06:01:06 avg_step=1.33s recent_step=1.37s eta=30:48:53 +2026-08-25 09:56:52 | INFO | train rank=0 | step=16350/100000 progress=16.4% loss=0.046585 lr=9.651919e-06 elapsed=06:01:22 avg_step=1.33s recent_step=1.57s eta=30:48:52 +2026-08-25 09:57:07 | INFO | train rank=0 | step=16360/100000 progress=16.4% loss=0.051773 lr=9.651313e-06 elapsed=06:01:37 avg_step=1.33s recent_step=1.53s eta=30:48:49 +2026-08-25 09:57:20 | INFO | train rank=0 | step=16370/100000 progress=16.4% loss=0.095459 lr=9.650706e-06 elapsed=06:01:50 avg_step=1.33s recent_step=1.22s eta=30:48:30 +2026-08-25 09:57:31 | INFO | train rank=0 | step=16380/100000 progress=16.4% loss=0.048393 lr=9.650099e-06 elapsed=06:02:01 avg_step=1.33s recent_step=1.18s eta=30:48:10 +2026-08-25 09:57:43 | INFO | train rank=0 | step=16390/100000 progress=16.4% loss=0.064392 lr=9.649491e-06 elapsed=06:02:13 avg_step=1.33s recent_step=1.14s eta=30:47:47 +2026-08-25 09:57:54 | INFO | train rank=0 | step=16400/100000 progress=16.4% loss=0.048828 lr=9.648882e-06 elapsed=06:02:24 avg_step=1.33s recent_step=1.16s eta=30:47:25 +2026-08-25 09:58:06 | INFO | train rank=0 | step=16410/100000 progress=16.4% loss=0.216797 lr=9.648273e-06 elapsed=06:02:36 avg_step=1.33s recent_step=1.13s eta=30:47:02 +2026-08-25 09:58:17 | INFO | train rank=0 | step=16420/100000 progress=16.4% loss=0.124756 lr=9.647664e-06 elapsed=06:02:47 avg_step=1.33s recent_step=1.14s eta=30:46:39 +2026-08-25 09:58:33 | INFO | train rank=0 | step=16430/100000 progress=16.4% loss=0.046722 lr=9.647054e-06 elapsed=06:03:03 avg_step=1.33s recent_step=1.55s eta=30:46:37 +2026-08-25 09:58:44 | INFO | train rank=0 | step=16440/100000 progress=16.4% loss=0.076202 lr=9.646444e-06 elapsed=06:03:14 avg_step=1.33s recent_step=1.18s eta=30:46:17 +2026-08-25 09:59:02 | INFO | train rank=0 | step=16450/100000 progress=16.4% loss=0.068848 lr=9.645833e-06 elapsed=06:03:32 avg_step=1.33s recent_step=1.79s eta=30:46:27 +2026-08-25 09:59:18 | INFO | train rank=0 | step=16460/100000 progress=16.5% loss=0.065918 lr=9.645221e-06 elapsed=06:03:48 avg_step=1.33s recent_step=1.58s eta=30:46:27 +2026-08-25 09:59:29 | INFO | train rank=0 | step=16470/100000 progress=16.5% loss=0.057892 lr=9.644609e-06 elapsed=06:03:59 avg_step=1.33s recent_step=1.10s eta=30:46:02 +2026-08-25 09:59:41 | INFO | train rank=0 | step=16480/100000 progress=16.5% loss=0.093201 lr=9.643997e-06 elapsed=06:04:11 avg_step=1.33s recent_step=1.17s eta=30:45:41 +2026-08-25 09:59:52 | INFO | train rank=0 | step=16490/100000 progress=16.5% loss=0.041161 lr=9.643384e-06 elapsed=06:04:22 avg_step=1.33s recent_step=1.15s eta=30:45:19 +2026-08-25 10:00:04 | INFO | train rank=0 | step=16500/100000 progress=16.5% loss=0.086212 lr=9.642770e-06 elapsed=06:04:34 avg_step=1.33s recent_step=1.16s eta=30:44:57 +2026-08-25 10:00:19 | INFO | train rank=0 | step=16510/100000 progress=16.5% loss=0.072144 lr=9.642156e-06 elapsed=06:04:48 avg_step=1.33s recent_step=1.45s eta=30:44:50 +2026-08-25 10:00:34 | INFO | train rank=0 | step=16520/100000 progress=16.5% loss=0.069366 lr=9.641542e-06 elapsed=06:05:04 avg_step=1.33s recent_step=1.54s eta=30:44:48 +2026-08-25 10:00:47 | INFO | train rank=0 | step=16530/100000 progress=16.5% loss=0.061646 lr=9.640927e-06 elapsed=06:05:16 avg_step=1.33s recent_step=1.27s eta=30:44:32 +2026-08-25 10:01:01 | INFO | train rank=0 | step=16540/100000 progress=16.5% loss=0.070190 lr=9.640311e-06 elapsed=06:05:30 avg_step=1.33s recent_step=1.39s eta=30:44:22 +2026-08-25 10:01:12 | INFO | train rank=0 | step=16550/100000 progress=16.6% loss=0.061462 lr=9.639695e-06 elapsed=06:05:42 avg_step=1.33s recent_step=1.19s eta=30:44:02 +2026-08-25 10:01:25 | INFO | train rank=0 | step=16560/100000 progress=16.6% loss=0.093445 lr=9.639078e-06 elapsed=06:05:55 avg_step=1.33s recent_step=1.23s eta=30:43:44 +2026-08-25 10:01:41 | INFO | train rank=0 | step=16570/100000 progress=16.6% loss=0.051300 lr=9.638461e-06 elapsed=06:06:11 avg_step=1.33s recent_step=1.67s eta=30:43:47 +2026-08-25 10:01:56 | INFO | train rank=0 | step=16580/100000 progress=16.6% loss=0.060486 lr=9.637844e-06 elapsed=06:06:25 avg_step=1.33s recent_step=1.41s eta=30:43:39 +2026-08-25 10:02:07 | INFO | train rank=0 | step=16590/100000 progress=16.6% loss=0.132385 lr=9.637226e-06 elapsed=06:06:37 avg_step=1.33s recent_step=1.14s eta=30:43:16 +2026-08-25 10:02:21 | INFO | train rank=0 | step=16600/100000 progress=16.6% loss=0.057678 lr=9.636607e-06 elapsed=06:06:51 avg_step=1.33s recent_step=1.45s eta=30:43:09 +2026-08-25 10:02:35 | INFO | train rank=0 | step=16610/100000 progress=16.6% loss=0.068054 lr=9.635988e-06 elapsed=06:07:05 avg_step=1.33s recent_step=1.41s eta=30:43:00 +2026-08-25 10:02:50 | INFO | train rank=0 | step=16620/100000 progress=16.6% loss=0.052086 lr=9.635368e-06 elapsed=06:07:20 avg_step=1.33s recent_step=1.43s eta=30:42:52 +2026-08-25 10:03:01 | INFO | train rank=0 | step=16630/100000 progress=16.6% loss=0.128723 lr=9.634748e-06 elapsed=06:07:31 avg_step=1.33s recent_step=1.15s eta=30:42:30 +2026-08-25 10:03:17 | INFO | train rank=0 | step=16640/100000 progress=16.6% loss=0.062653 lr=9.634128e-06 elapsed=06:07:47 avg_step=1.33s recent_step=1.59s eta=30:42:30 +2026-08-25 10:03:29 | INFO | train rank=0 | step=16650/100000 progress=16.7% loss=0.074753 lr=9.633507e-06 elapsed=06:07:58 avg_step=1.33s recent_step=1.14s eta=30:42:07 +2026-08-25 10:03:43 | INFO | train rank=0 | step=16660/100000 progress=16.7% loss=0.055649 lr=9.632885e-06 elapsed=06:08:13 avg_step=1.33s recent_step=1.45s eta=30:42:00 +2026-08-25 10:03:55 | INFO | train rank=0 | step=16670/100000 progress=16.7% loss=0.084198 lr=9.632263e-06 elapsed=06:08:25 avg_step=1.33s recent_step=1.19s eta=30:41:40 +2026-08-25 10:04:09 | INFO | train rank=0 | step=16680/100000 progress=16.7% loss=0.063889 lr=9.631640e-06 elapsed=06:08:39 avg_step=1.33s recent_step=1.44s eta=30:41:32 +2026-08-25 10:04:22 | INFO | train rank=0 | step=16690/100000 progress=16.7% loss=0.126221 lr=9.631017e-06 elapsed=06:08:52 avg_step=1.33s recent_step=1.26s eta=30:41:15 +2026-08-25 10:04:34 | INFO | train rank=0 | step=16700/100000 progress=16.7% loss=0.062073 lr=9.630393e-06 elapsed=06:09:04 avg_step=1.33s recent_step=1.18s eta=30:40:55 +2026-08-25 10:04:46 | INFO | train rank=0 | step=16710/100000 progress=16.7% loss=0.107605 lr=9.629769e-06 elapsed=06:09:16 avg_step=1.33s recent_step=1.24s eta=30:40:38 +2026-08-25 10:04:59 | INFO | train rank=0 | step=16720/100000 progress=16.7% loss=0.056030 lr=9.629145e-06 elapsed=06:09:29 avg_step=1.33s recent_step=1.32s eta=30:40:24 +2026-08-25 10:05:13 | INFO | train rank=0 | step=16730/100000 progress=16.7% loss=0.069305 lr=9.628519e-06 elapsed=06:09:43 avg_step=1.33s recent_step=1.39s eta=30:40:14 +2026-08-25 10:05:25 | INFO | train rank=0 | step=16740/100000 progress=16.7% loss=0.046661 lr=9.627894e-06 elapsed=06:09:54 avg_step=1.33s recent_step=1.13s eta=30:39:51 +2026-08-25 10:05:37 | INFO | train rank=0 | step=16750/100000 progress=16.8% loss=0.049789 lr=9.627268e-06 elapsed=06:10:06 avg_step=1.33s recent_step=1.20s eta=30:39:31 +2026-08-25 10:05:53 | INFO | train rank=0 | step=16760/100000 progress=16.8% loss=0.054504 lr=9.626641e-06 elapsed=06:10:22 avg_step=1.33s recent_step=1.59s eta=30:39:32 +2026-08-25 10:06:06 | INFO | train rank=0 | step=16770/100000 progress=16.8% loss=0.064362 lr=9.626014e-06 elapsed=06:10:36 avg_step=1.33s recent_step=1.37s eta=30:39:20 +2026-08-25 10:06:21 | INFO | train rank=0 | step=16780/100000 progress=16.8% loss=0.053772 lr=9.625386e-06 elapsed=06:10:51 avg_step=1.33s recent_step=1.50s eta=30:39:16 +2026-08-25 10:06:33 | INFO | train rank=0 | step=16790/100000 progress=16.8% loss=0.057495 lr=9.624758e-06 elapsed=06:11:03 avg_step=1.33s recent_step=1.19s eta=30:38:56 +2026-08-25 10:06:50 | INFO | train rank=0 | step=16800/100000 progress=16.8% loss=0.056641 lr=9.624129e-06 elapsed=06:11:20 avg_step=1.33s recent_step=1.70s eta=30:39:01 +2026-08-25 10:07:05 | INFO | train rank=0 | step=16810/100000 progress=16.8% loss=0.058441 lr=9.623500e-06 elapsed=06:11:35 avg_step=1.33s recent_step=1.51s eta=30:38:57 +2026-08-25 10:07:18 | INFO | train rank=0 | step=16820/100000 progress=16.8% loss=0.058044 lr=9.622870e-06 elapsed=06:11:48 avg_step=1.33s recent_step=1.31s eta=30:38:43 +2026-08-25 10:07:33 | INFO | train rank=0 | step=16830/100000 progress=16.8% loss=0.065445 lr=9.622240e-06 elapsed=06:12:03 avg_step=1.33s recent_step=1.51s eta=30:38:39 +2026-08-25 10:07:45 | INFO | train rank=0 | step=16840/100000 progress=16.8% loss=0.074768 lr=9.621609e-06 elapsed=06:12:15 avg_step=1.33s recent_step=1.16s eta=30:38:17 +2026-08-25 10:07:57 | INFO | train rank=0 | step=16850/100000 progress=16.9% loss=0.064209 lr=9.620978e-06 elapsed=06:12:27 avg_step=1.33s recent_step=1.18s eta=30:37:57 +2026-08-25 10:08:09 | INFO | train rank=0 | step=16860/100000 progress=16.9% loss=0.058090 lr=9.620346e-06 elapsed=06:12:39 avg_step=1.33s recent_step=1.22s eta=30:37:38 +2026-08-25 10:08:20 | INFO | train rank=0 | step=16870/100000 progress=16.9% loss=0.035561 lr=9.619714e-06 elapsed=06:12:50 avg_step=1.33s recent_step=1.15s eta=30:37:16 +2026-08-25 10:08:33 | INFO | train rank=0 | step=16880/100000 progress=16.9% loss=0.072327 lr=9.619081e-06 elapsed=06:13:03 avg_step=1.33s recent_step=1.30s eta=30:37:01 +2026-08-25 10:08:48 | INFO | train rank=0 | step=16890/100000 progress=16.9% loss=0.090851 lr=9.618448e-06 elapsed=06:13:18 avg_step=1.33s recent_step=1.43s eta=30:36:53 +2026-08-25 10:08:59 | INFO | train rank=0 | step=16900/100000 progress=16.9% loss=0.054893 lr=9.617814e-06 elapsed=06:13:29 avg_step=1.33s recent_step=1.15s eta=30:36:31 +2026-08-25 10:09:14 | INFO | train rank=0 | step=16910/100000 progress=16.9% loss=0.062744 lr=9.617180e-06 elapsed=06:13:44 avg_step=1.33s recent_step=1.52s eta=30:36:27 +2026-08-25 10:09:26 | INFO | train rank=0 | step=16920/100000 progress=16.9% loss=0.097961 lr=9.616545e-06 elapsed=06:13:56 avg_step=1.33s recent_step=1.20s eta=30:36:08 +2026-08-25 10:09:39 | INFO | train rank=0 | step=16930/100000 progress=16.9% loss=0.075043 lr=9.615910e-06 elapsed=06:14:09 avg_step=1.33s recent_step=1.31s eta=30:35:54 +2026-08-25 10:09:54 | INFO | train rank=0 | step=16940/100000 progress=16.9% loss=0.045441 lr=9.615274e-06 elapsed=06:14:24 avg_step=1.33s recent_step=1.50s eta=30:35:49 +2026-08-25 10:10:06 | INFO | train rank=0 | step=16950/100000 progress=17.0% loss=0.070526 lr=9.614638e-06 elapsed=06:14:36 avg_step=1.33s recent_step=1.15s eta=30:35:27 +2026-08-25 10:10:19 | INFO | train rank=0 | step=16960/100000 progress=17.0% loss=0.063568 lr=9.614001e-06 elapsed=06:14:49 avg_step=1.33s recent_step=1.29s eta=30:35:12 +2026-08-25 10:10:35 | INFO | train rank=0 | step=16970/100000 progress=17.0% loss=0.085938 lr=9.613364e-06 elapsed=06:15:05 avg_step=1.33s recent_step=1.64s eta=30:35:14 +2026-08-25 10:10:50 | INFO | train rank=0 | step=16980/100000 progress=17.0% loss=0.051697 lr=9.612726e-06 elapsed=06:15:20 avg_step=1.33s recent_step=1.49s eta=30:35:09 +2026-08-25 10:11:02 | INFO | train rank=0 | step=16990/100000 progress=17.0% loss=0.087341 lr=9.612088e-06 elapsed=06:15:32 avg_step=1.33s recent_step=1.16s eta=30:34:47 +2026-08-25 10:11:19 | INFO | train rank=0 | step=17000/100000 progress=17.0% loss=0.068604 lr=9.611449e-06 elapsed=06:15:49 avg_step=1.33s recent_step=1.75s eta=30:34:55 +2026-08-25 10:11:31 | INFO | train rank=0 | step=17010/100000 progress=17.0% loss=0.064667 lr=9.610809e-06 elapsed=06:16:01 avg_step=1.33s recent_step=1.20s eta=30:34:36 +2026-08-25 10:11:43 | INFO | train rank=0 | step=17020/100000 progress=17.0% loss=0.070587 lr=9.610170e-06 elapsed=06:16:13 avg_step=1.33s recent_step=1.15s eta=30:34:14 +2026-08-25 10:12:01 | INFO | train rank=0 | step=17030/100000 progress=17.0% loss=0.038727 lr=9.609529e-06 elapsed=06:16:31 avg_step=1.33s recent_step=1.80s eta=30:34:23 +2026-08-25 10:12:17 | INFO | train rank=0 | step=17040/100000 progress=17.0% loss=0.053925 lr=9.608888e-06 elapsed=06:16:47 avg_step=1.33s recent_step=1.59s eta=30:34:23 +2026-08-25 10:12:28 | INFO | train rank=0 | step=17050/100000 progress=17.1% loss=0.049927 lr=9.608247e-06 elapsed=06:16:58 avg_step=1.33s recent_step=1.13s eta=30:34:00 +2026-08-25 10:12:40 | INFO | train rank=0 | step=17060/100000 progress=17.1% loss=0.100494 lr=9.607605e-06 elapsed=06:17:10 avg_step=1.33s recent_step=1.18s eta=30:33:40 +2026-08-25 10:12:51 | INFO | train rank=0 | step=17070/100000 progress=17.1% loss=0.042587 lr=9.606963e-06 elapsed=06:17:21 avg_step=1.33s recent_step=1.15s eta=30:33:18 +2026-08-25 10:13:04 | INFO | train rank=0 | step=17080/100000 progress=17.1% loss=0.078766 lr=9.606320e-06 elapsed=06:17:34 avg_step=1.33s recent_step=1.25s eta=30:33:01 +2026-08-25 10:13:19 | INFO | train rank=0 | step=17090/100000 progress=17.1% loss=0.056122 lr=9.605677e-06 elapsed=06:17:49 avg_step=1.33s recent_step=1.50s eta=30:32:56 +2026-08-25 10:13:30 | INFO | train rank=0 | step=17100/100000 progress=17.1% loss=0.074493 lr=9.605033e-06 elapsed=06:18:00 avg_step=1.33s recent_step=1.14s eta=30:32:34 +2026-08-25 10:13:43 | INFO | train rank=0 | step=17110/100000 progress=17.1% loss=0.052338 lr=9.604389e-06 elapsed=06:18:13 avg_step=1.33s recent_step=1.28s eta=30:32:18 +2026-08-25 10:13:57 | INFO | train rank=0 | step=17120/100000 progress=17.1% loss=0.055038 lr=9.603744e-06 elapsed=06:18:27 avg_step=1.33s recent_step=1.40s eta=30:32:09 +2026-08-25 10:14:08 | INFO | train rank=0 | step=17130/100000 progress=17.1% loss=0.042862 lr=9.603098e-06 elapsed=06:18:38 avg_step=1.33s recent_step=1.10s eta=30:31:45 +2026-08-25 10:14:22 | INFO | train rank=0 | step=17140/100000 progress=17.1% loss=0.066101 lr=9.602452e-06 elapsed=06:18:52 avg_step=1.33s recent_step=1.41s eta=30:31:36 +2026-08-25 10:14:34 | INFO | train rank=0 | step=17150/100000 progress=17.2% loss=0.089661 lr=9.601806e-06 elapsed=06:19:04 avg_step=1.33s recent_step=1.21s eta=30:31:17 +2026-08-25 10:14:50 | INFO | train rank=0 | step=17160/100000 progress=17.2% loss=0.060043 lr=9.601159e-06 elapsed=06:19:19 avg_step=1.33s recent_step=1.53s eta=30:31:13 +2026-08-25 10:15:02 | INFO | train rank=0 | step=17170/100000 progress=17.2% loss=0.077026 lr=9.600512e-06 elapsed=06:19:32 avg_step=1.33s recent_step=1.28s eta=30:30:58 +2026-08-25 10:15:19 | INFO | train rank=0 | step=17180/100000 progress=17.2% loss=0.061783 lr=9.599864e-06 elapsed=06:19:49 avg_step=1.33s recent_step=1.65s eta=30:31:00 +2026-08-25 10:15:33 | INFO | train rank=0 | step=17190/100000 progress=17.2% loss=0.050751 lr=9.599216e-06 elapsed=06:20:03 avg_step=1.33s recent_step=1.38s eta=30:30:50 +2026-08-25 10:15:44 | INFO | train rank=0 | step=17200/100000 progress=17.2% loss=0.065338 lr=9.598567e-06 elapsed=06:20:14 avg_step=1.33s recent_step=1.16s eta=30:30:28 +2026-08-25 10:15:57 | INFO | train rank=0 | step=17210/100000 progress=17.2% loss=0.071106 lr=9.597917e-06 elapsed=06:20:27 avg_step=1.33s recent_step=1.24s eta=30:30:11 +2026-08-25 10:16:09 | INFO | train rank=0 | step=17220/100000 progress=17.2% loss=0.090027 lr=9.597267e-06 elapsed=06:20:39 avg_step=1.33s recent_step=1.19s eta=30:29:51 +2026-08-25 10:16:23 | INFO | train rank=0 | step=17230/100000 progress=17.2% loss=0.070618 lr=9.596617e-06 elapsed=06:20:53 avg_step=1.33s recent_step=1.48s eta=30:29:46 +2026-08-25 10:16:37 | INFO | train rank=0 | step=17240/100000 progress=17.2% loss=0.080109 lr=9.595966e-06 elapsed=06:21:07 avg_step=1.33s recent_step=1.39s eta=30:29:35 +2026-08-25 10:16:50 | INFO | train rank=0 | step=17250/100000 progress=17.2% loss=0.047867 lr=9.595315e-06 elapsed=06:21:20 avg_step=1.33s recent_step=1.31s eta=30:29:21 +2026-08-25 10:17:06 | INFO | train rank=0 | step=17260/100000 progress=17.3% loss=0.050140 lr=9.594663e-06 elapsed=06:21:36 avg_step=1.33s recent_step=1.53s eta=30:29:17 +2026-08-25 10:17:17 | INFO | train rank=0 | step=17270/100000 progress=17.3% loss=0.068726 lr=9.594010e-06 elapsed=06:21:47 avg_step=1.33s recent_step=1.15s eta=30:28:56 +2026-08-25 10:17:32 | INFO | train rank=0 | step=17280/100000 progress=17.3% loss=0.122803 lr=9.593358e-06 elapsed=06:22:02 avg_step=1.33s recent_step=1.52s eta=30:28:52 +2026-08-25 10:17:48 | INFO | train rank=0 | step=17290/100000 progress=17.3% loss=0.045822 lr=9.592704e-06 elapsed=06:22:17 avg_step=1.33s recent_step=1.52s eta=30:28:48 +2026-08-25 10:18:00 | INFO | train rank=0 | step=17300/100000 progress=17.3% loss=0.070831 lr=9.592050e-06 elapsed=06:22:30 avg_step=1.33s recent_step=1.23s eta=30:28:30 +2026-08-25 10:18:14 | INFO | train rank=0 | step=17310/100000 progress=17.3% loss=0.087952 lr=9.591396e-06 elapsed=06:22:44 avg_step=1.33s recent_step=1.44s eta=30:28:22 +2026-08-25 10:18:27 | INFO | train rank=0 | step=17320/100000 progress=17.3% loss=0.070679 lr=9.590741e-06 elapsed=06:22:57 avg_step=1.33s recent_step=1.30s eta=30:28:08 +2026-08-25 10:18:43 | INFO | train rank=0 | step=17330/100000 progress=17.3% loss=0.067444 lr=9.590085e-06 elapsed=06:23:12 avg_step=1.33s recent_step=1.51s eta=30:28:03 +2026-08-25 10:18:56 | INFO | train rank=0 | step=17340/100000 progress=17.3% loss=0.067810 lr=9.589430e-06 elapsed=06:23:25 avg_step=1.33s recent_step=1.30s eta=30:27:49 +2026-08-25 10:19:07 | INFO | train rank=0 | step=17350/100000 progress=17.3% loss=0.048355 lr=9.588773e-06 elapsed=06:23:37 avg_step=1.33s recent_step=1.19s eta=30:27:29 +2026-08-25 10:19:21 | INFO | train rank=0 | step=17360/100000 progress=17.4% loss=0.066559 lr=9.588116e-06 elapsed=06:23:51 avg_step=1.33s recent_step=1.38s eta=30:27:19 +2026-08-25 10:19:35 | INFO | train rank=0 | step=17370/100000 progress=17.4% loss=0.047287 lr=9.587459e-06 elapsed=06:24:05 avg_step=1.33s recent_step=1.35s eta=30:27:07 +2026-08-25 10:19:47 | INFO | train rank=0 | step=17380/100000 progress=17.4% loss=0.058228 lr=9.586801e-06 elapsed=06:24:17 avg_step=1.33s recent_step=1.21s eta=30:26:48 +2026-08-25 10:19:59 | INFO | train rank=0 | step=17390/100000 progress=17.4% loss=0.092590 lr=9.586142e-06 elapsed=06:24:29 avg_step=1.33s recent_step=1.18s eta=30:26:28 +2026-08-25 10:20:11 | INFO | train rank=0 | step=17400/100000 progress=17.4% loss=0.043182 lr=9.585483e-06 elapsed=06:24:40 avg_step=1.33s recent_step=1.18s eta=30:26:08 +2026-08-25 10:20:24 | INFO | train rank=0 | step=17410/100000 progress=17.4% loss=0.042175 lr=9.584824e-06 elapsed=06:24:54 avg_step=1.33s recent_step=1.36s eta=30:25:56 +2026-08-25 10:20:37 | INFO | train rank=0 | step=17420/100000 progress=17.4% loss=0.063385 lr=9.584164e-06 elapsed=06:25:07 avg_step=1.33s recent_step=1.25s eta=30:25:39 +2026-08-25 10:20:50 | INFO | train rank=0 | step=17430/100000 progress=17.4% loss=0.054153 lr=9.583504e-06 elapsed=06:25:20 avg_step=1.33s recent_step=1.32s eta=30:25:25 +2026-08-25 10:21:04 | INFO | train rank=0 | step=17440/100000 progress=17.4% loss=0.077393 lr=9.582843e-06 elapsed=06:25:33 avg_step=1.33s recent_step=1.37s eta=30:25:14 +2026-08-25 10:21:15 | INFO | train rank=0 | step=17450/100000 progress=17.4% loss=0.052429 lr=9.582181e-06 elapsed=06:25:45 avg_step=1.33s recent_step=1.13s eta=30:24:52 +2026-08-25 10:21:29 | INFO | train rank=0 | step=17460/100000 progress=17.5% loss=0.055161 lr=9.581519e-06 elapsed=06:25:59 avg_step=1.33s recent_step=1.38s eta=30:24:41 +2026-08-25 10:21:41 | INFO | train rank=0 | step=17470/100000 progress=17.5% loss=0.079529 lr=9.580857e-06 elapsed=06:26:11 avg_step=1.33s recent_step=1.25s eta=30:24:24 +2026-08-25 10:21:53 | INFO | train rank=0 | step=17480/100000 progress=17.5% loss=0.041801 lr=9.580194e-06 elapsed=06:26:23 avg_step=1.33s recent_step=1.17s eta=30:24:03 +2026-08-25 10:22:06 | INFO | train rank=0 | step=17490/100000 progress=17.5% loss=0.075531 lr=9.579531e-06 elapsed=06:26:36 avg_step=1.33s recent_step=1.36s eta=30:23:52 +2026-08-25 10:22:19 | INFO | train rank=0 | step=17500/100000 progress=17.5% loss=0.055420 lr=9.578867e-06 elapsed=06:26:49 avg_step=1.33s recent_step=1.27s eta=30:23:35 +2026-08-25 10:22:33 | INFO | train rank=0 | step=17510/100000 progress=17.5% loss=0.059601 lr=9.578202e-06 elapsed=06:27:02 avg_step=1.33s recent_step=1.35s eta=30:23:23 +2026-08-25 10:22:46 | INFO | train rank=0 | step=17520/100000 progress=17.5% loss=0.077759 lr=9.577537e-06 elapsed=06:27:16 avg_step=1.33s recent_step=1.34s eta=30:23:11 +2026-08-25 10:23:01 | INFO | train rank=0 | step=17530/100000 progress=17.5% loss=0.041061 lr=9.576872e-06 elapsed=06:27:31 avg_step=1.33s recent_step=1.48s eta=30:23:05 +2026-08-25 10:23:13 | INFO | train rank=0 | step=17540/100000 progress=17.5% loss=0.048676 lr=9.576206e-06 elapsed=06:27:43 avg_step=1.33s recent_step=1.24s eta=30:22:47 +2026-08-25 10:23:30 | INFO | train rank=0 | step=17550/100000 progress=17.5% loss=0.080032 lr=9.575539e-06 elapsed=06:28:00 avg_step=1.33s recent_step=1.65s eta=30:22:49 +2026-08-25 10:23:42 | INFO | train rank=0 | step=17560/100000 progress=17.6% loss=0.089905 lr=9.574872e-06 elapsed=06:28:12 avg_step=1.33s recent_step=1.20s eta=30:22:30 +2026-08-25 10:23:57 | INFO | train rank=0 | step=17570/100000 progress=17.6% loss=0.039894 lr=9.574205e-06 elapsed=06:28:26 avg_step=1.33s recent_step=1.50s eta=30:22:25 +2026-08-25 10:24:08 | INFO | train rank=0 | step=17580/100000 progress=17.6% loss=0.133057 lr=9.573537e-06 elapsed=06:28:38 avg_step=1.33s recent_step=1.10s eta=30:22:01 +2026-08-25 10:24:20 | INFO | train rank=0 | step=17590/100000 progress=17.6% loss=0.046555 lr=9.572869e-06 elapsed=06:28:50 avg_step=1.33s recent_step=1.24s eta=30:21:43 +2026-08-25 10:24:37 | INFO | train rank=0 | step=17600/100000 progress=17.6% loss=0.078033 lr=9.572200e-06 elapsed=06:29:07 avg_step=1.33s recent_step=1.73s eta=30:21:49 +2026-08-25 10:24:52 | INFO | train rank=0 | step=17610/100000 progress=17.6% loss=0.046295 lr=9.571530e-06 elapsed=06:29:22 avg_step=1.33s recent_step=1.47s eta=30:21:42 +2026-08-25 10:25:10 | INFO | train rank=0 | step=17620/100000 progress=17.6% loss=0.077362 lr=9.570860e-06 elapsed=06:29:40 avg_step=1.33s recent_step=1.82s eta=30:21:52 +2026-08-25 10:25:21 | INFO | train rank=0 | step=17630/100000 progress=17.6% loss=0.095703 lr=9.570190e-06 elapsed=06:29:51 avg_step=1.33s recent_step=1.10s eta=30:21:28 +2026-08-25 10:25:35 | INFO | train rank=0 | step=17640/100000 progress=17.6% loss=0.075806 lr=9.569519e-06 elapsed=06:30:05 avg_step=1.33s recent_step=1.43s eta=30:21:20 +2026-08-25 10:25:48 | INFO | train rank=0 | step=17650/100000 progress=17.6% loss=0.070343 lr=9.568847e-06 elapsed=06:30:18 avg_step=1.33s recent_step=1.23s eta=30:21:02 +2026-08-25 10:26:01 | INFO | train rank=0 | step=17660/100000 progress=17.7% loss=0.069885 lr=9.568176e-06 elapsed=06:30:31 avg_step=1.33s recent_step=1.31s eta=30:20:48 +2026-08-25 10:26:16 | INFO | train rank=0 | step=17670/100000 progress=17.7% loss=0.079010 lr=9.567503e-06 elapsed=06:30:46 avg_step=1.33s recent_step=1.56s eta=30:20:45 +2026-08-25 10:26:28 | INFO | train rank=0 | step=17680/100000 progress=17.7% loss=0.058197 lr=9.566830e-06 elapsed=06:30:58 avg_step=1.33s recent_step=1.15s eta=30:20:23 +2026-08-25 10:26:39 | INFO | train rank=0 | step=17690/100000 progress=17.7% loss=0.056656 lr=9.566157e-06 elapsed=06:31:09 avg_step=1.33s recent_step=1.14s eta=30:20:01 +2026-08-25 10:26:51 | INFO | train rank=0 | step=17700/100000 progress=17.7% loss=0.088287 lr=9.565483e-06 elapsed=06:31:21 avg_step=1.33s recent_step=1.18s eta=30:19:41 +2026-08-25 10:27:03 | INFO | train rank=0 | step=17710/100000 progress=17.7% loss=0.069916 lr=9.564808e-06 elapsed=06:31:33 avg_step=1.33s recent_step=1.20s eta=30:19:22 +2026-08-25 10:27:15 | INFO | train rank=0 | step=17720/100000 progress=17.7% loss=0.046982 lr=9.564133e-06 elapsed=06:31:45 avg_step=1.33s recent_step=1.21s eta=30:19:04 +2026-08-25 10:27:27 | INFO | train rank=0 | step=17730/100000 progress=17.7% loss=0.054413 lr=9.563458e-06 elapsed=06:31:57 avg_step=1.33s recent_step=1.16s eta=30:18:42 +2026-08-25 10:27:38 | INFO | train rank=0 | step=17740/100000 progress=17.7% loss=0.055969 lr=9.562782e-06 elapsed=06:32:08 avg_step=1.33s recent_step=1.15s eta=30:18:21 +2026-08-25 10:27:50 | INFO | train rank=0 | step=17750/100000 progress=17.8% loss=0.074219 lr=9.562106e-06 elapsed=06:32:20 avg_step=1.33s recent_step=1.17s eta=30:18:00 +2026-08-25 10:28:01 | INFO | train rank=0 | step=17760/100000 progress=17.8% loss=0.075836 lr=9.561429e-06 elapsed=06:32:31 avg_step=1.33s recent_step=1.14s eta=30:17:39 +2026-08-25 10:28:13 | INFO | train rank=0 | step=17770/100000 progress=17.8% loss=0.055237 lr=9.560751e-06 elapsed=06:32:43 avg_step=1.33s recent_step=1.18s eta=30:17:19 +2026-08-25 10:28:25 | INFO | train rank=0 | step=17780/100000 progress=17.8% loss=0.043747 lr=9.560073e-06 elapsed=06:32:55 avg_step=1.33s recent_step=1.21s eta=30:17:00 +2026-08-25 10:28:39 | INFO | train rank=0 | step=17790/100000 progress=17.8% loss=0.056656 lr=9.559395e-06 elapsed=06:33:09 avg_step=1.33s recent_step=1.35s eta=30:16:48 +2026-08-25 10:28:52 | INFO | train rank=0 | step=17800/100000 progress=17.8% loss=0.113342 lr=9.558716e-06 elapsed=06:33:22 avg_step=1.33s recent_step=1.37s eta=30:16:37 +2026-08-25 10:29:08 | INFO | train rank=0 | step=17810/100000 progress=17.8% loss=0.050308 lr=9.558036e-06 elapsed=06:33:38 avg_step=1.33s recent_step=1.53s eta=30:16:33 +2026-08-25 10:29:25 | INFO | train rank=0 | step=17820/100000 progress=17.8% loss=0.076294 lr=9.557357e-06 elapsed=06:33:55 avg_step=1.33s recent_step=1.70s eta=30:16:37 +2026-08-25 10:29:41 | INFO | train rank=0 | step=17830/100000 progress=17.8% loss=0.062408 lr=9.556676e-06 elapsed=06:34:11 avg_step=1.33s recent_step=1.62s eta=30:16:37 +2026-08-25 10:29:53 | INFO | train rank=0 | step=17840/100000 progress=17.8% loss=0.052307 lr=9.555995e-06 elapsed=06:34:23 avg_step=1.33s recent_step=1.18s eta=30:16:17 +2026-08-25 10:30:08 | INFO | train rank=0 | step=17850/100000 progress=17.8% loss=0.063080 lr=9.555314e-06 elapsed=06:34:38 avg_step=1.33s recent_step=1.52s eta=30:16:13 +2026-08-25 10:30:20 | INFO | train rank=0 | step=17860/100000 progress=17.9% loss=0.040817 lr=9.554632e-06 elapsed=06:34:50 avg_step=1.33s recent_step=1.17s eta=30:15:52 +2026-08-25 10:30:36 | INFO | train rank=0 | step=17870/100000 progress=17.9% loss=0.046204 lr=9.553949e-06 elapsed=06:35:06 avg_step=1.33s recent_step=1.65s eta=30:15:54 +2026-08-25 10:30:48 | INFO | train rank=0 | step=17880/100000 progress=17.9% loss=0.065613 lr=9.553267e-06 elapsed=06:35:18 avg_step=1.33s recent_step=1.19s eta=30:15:35 +2026-08-25 10:31:04 | INFO | train rank=0 | step=17890/100000 progress=17.9% loss=0.058655 lr=9.552583e-06 elapsed=06:35:34 avg_step=1.33s recent_step=1.60s eta=30:15:34 +2026-08-25 10:31:22 | INFO | train rank=0 | step=17900/100000 progress=17.9% loss=0.049530 lr=9.551899e-06 elapsed=06:35:52 avg_step=1.33s recent_step=1.76s eta=30:15:41 +2026-08-25 10:31:33 | INFO | train rank=0 | step=17910/100000 progress=17.9% loss=0.063568 lr=9.551215e-06 elapsed=06:36:03 avg_step=1.33s recent_step=1.16s eta=30:15:20 +2026-08-25 10:31:45 | INFO | train rank=0 | step=17920/100000 progress=17.9% loss=0.082428 lr=9.550530e-06 elapsed=06:36:14 avg_step=1.33s recent_step=1.13s eta=30:14:57 +2026-08-25 10:32:02 | INFO | train rank=0 | step=17930/100000 progress=17.9% loss=0.070496 lr=9.549844e-06 elapsed=06:36:32 avg_step=1.33s recent_step=1.73s eta=30:15:03 +2026-08-25 10:32:15 | INFO | train rank=0 | step=17940/100000 progress=17.9% loss=0.049316 lr=9.549159e-06 elapsed=06:36:45 avg_step=1.33s recent_step=1.34s eta=30:14:50 +2026-08-25 10:32:27 | INFO | train rank=0 | step=17950/100000 progress=17.9% loss=0.061005 lr=9.548472e-06 elapsed=06:36:57 avg_step=1.33s recent_step=1.13s eta=30:14:28 +2026-08-25 10:32:42 | INFO | train rank=0 | step=17960/100000 progress=18.0% loss=0.074524 lr=9.547785e-06 elapsed=06:37:12 avg_step=1.33s recent_step=1.52s eta=30:14:23 +2026-08-25 10:32:53 | INFO | train rank=0 | step=17970/100000 progress=18.0% loss=0.039825 lr=9.547098e-06 elapsed=06:37:23 avg_step=1.33s recent_step=1.13s eta=30:14:01 +2026-08-25 10:33:09 | INFO | train rank=0 | step=17980/100000 progress=18.0% loss=0.048721 lr=9.546410e-06 elapsed=06:37:39 avg_step=1.33s recent_step=1.58s eta=30:14:00 +2026-08-25 10:33:23 | INFO | train rank=0 | step=17990/100000 progress=18.0% loss=0.057678 lr=9.545722e-06 elapsed=06:37:53 avg_step=1.33s recent_step=1.41s eta=30:13:50 +2026-08-25 10:33:35 | INFO | train rank=0 | step=18000/100000 progress=18.0% loss=0.044403 lr=9.545033e-06 elapsed=06:38:04 avg_step=1.33s recent_step=1.15s eta=30:13:29 +2026-08-25 10:33:47 | INFO | train rank=0 | step=18010/100000 progress=18.0% loss=0.058151 lr=9.544343e-06 elapsed=06:38:16 avg_step=1.33s recent_step=1.20s eta=30:13:10 +2026-08-25 10:33:58 | INFO | train rank=0 | step=18020/100000 progress=18.0% loss=0.077698 lr=9.543653e-06 elapsed=06:38:28 avg_step=1.33s recent_step=1.18s eta=30:12:50 +2026-08-25 10:34:10 | INFO | train rank=0 | step=18030/100000 progress=18.0% loss=0.051132 lr=9.542963e-06 elapsed=06:38:40 avg_step=1.33s recent_step=1.18s eta=30:12:30 +2026-08-25 10:34:26 | INFO | train rank=0 | step=18040/100000 progress=18.0% loss=0.069855 lr=9.542272e-06 elapsed=06:38:56 avg_step=1.33s recent_step=1.54s eta=30:12:27 +2026-08-25 10:34:37 | INFO | train rank=0 | step=18050/100000 progress=18.1% loss=0.041672 lr=9.541581e-06 elapsed=06:39:07 avg_step=1.33s recent_step=1.13s eta=30:12:05 +2026-08-25 10:34:49 | INFO | train rank=0 | step=18060/100000 progress=18.1% loss=0.058243 lr=9.540889e-06 elapsed=06:39:19 avg_step=1.33s recent_step=1.21s eta=30:11:46 +2026-08-25 10:35:01 | INFO | train rank=0 | step=18070/100000 progress=18.1% loss=0.049988 lr=9.540197e-06 elapsed=06:39:31 avg_step=1.33s recent_step=1.17s eta=30:11:25 +2026-08-25 10:35:14 | INFO | train rank=0 | step=18080/100000 progress=18.1% loss=0.070190 lr=9.539504e-06 elapsed=06:39:44 avg_step=1.33s recent_step=1.37s eta=30:11:14 +2026-08-25 10:35:26 | INFO | train rank=0 | step=18090/100000 progress=18.1% loss=0.062164 lr=9.538810e-06 elapsed=06:39:56 avg_step=1.33s recent_step=1.18s eta=30:10:54 +2026-08-25 10:35:39 | INFO | train rank=0 | step=18100/100000 progress=18.1% loss=0.056610 lr=9.538116e-06 elapsed=06:40:09 avg_step=1.33s recent_step=1.27s eta=30:10:38 +2026-08-25 10:35:51 | INFO | train rank=0 | step=18110/100000 progress=18.1% loss=0.059631 lr=9.537422e-06 elapsed=06:40:21 avg_step=1.33s recent_step=1.24s eta=30:10:21 +2026-08-25 10:36:08 | INFO | train rank=0 | step=18120/100000 progress=18.1% loss=0.065628 lr=9.536727e-06 elapsed=06:40:38 avg_step=1.33s recent_step=1.67s eta=30:10:23 +2026-08-25 10:36:19 | INFO | train rank=0 | step=18130/100000 progress=18.1% loss=0.088013 lr=9.536032e-06 elapsed=06:40:49 avg_step=1.33s recent_step=1.13s eta=30:10:01 +2026-08-25 10:36:35 | INFO | train rank=0 | step=18140/100000 progress=18.1% loss=0.064087 lr=9.535336e-06 elapsed=06:41:04 avg_step=1.33s recent_step=1.53s eta=30:09:57 +2026-08-25 10:36:48 | INFO | train rank=0 | step=18150/100000 progress=18.1% loss=0.051178 lr=9.534640e-06 elapsed=06:41:18 avg_step=1.33s recent_step=1.38s eta=30:09:46 +2026-08-25 10:37:02 | INFO | train rank=0 | step=18160/100000 progress=18.2% loss=0.079788 lr=9.533943e-06 elapsed=06:41:32 avg_step=1.33s recent_step=1.39s eta=30:09:36 +2026-08-25 10:37:14 | INFO | train rank=0 | step=18170/100000 progress=18.2% loss=0.050873 lr=9.533246e-06 elapsed=06:41:44 avg_step=1.33s recent_step=1.16s eta=30:09:15 +2026-08-25 10:37:26 | INFO | train rank=0 | step=18180/100000 progress=18.2% loss=0.054779 lr=9.532548e-06 elapsed=06:41:56 avg_step=1.33s recent_step=1.18s eta=30:08:55 +2026-08-25 10:37:42 | INFO | train rank=0 | step=18190/100000 progress=18.2% loss=0.048920 lr=9.531849e-06 elapsed=06:42:12 avg_step=1.33s recent_step=1.68s eta=30:08:58 +2026-08-25 10:37:54 | INFO | train rank=0 | step=18200/100000 progress=18.2% loss=0.104614 lr=9.531151e-06 elapsed=06:42:24 avg_step=1.33s recent_step=1.18s eta=30:08:38 +2026-08-25 10:38:07 | INFO | train rank=0 | step=18210/100000 progress=18.2% loss=0.082214 lr=9.530451e-06 elapsed=06:42:37 avg_step=1.33s recent_step=1.25s eta=30:08:21 +2026-08-25 10:38:20 | INFO | train rank=0 | step=18220/100000 progress=18.2% loss=0.069489 lr=9.529752e-06 elapsed=06:42:50 avg_step=1.33s recent_step=1.29s eta=30:08:06 +2026-08-25 10:38:35 | INFO | train rank=0 | step=18230/100000 progress=18.2% loss=0.077911 lr=9.529051e-06 elapsed=06:43:05 avg_step=1.33s recent_step=1.55s eta=30:08:03 +2026-08-25 10:38:50 | INFO | train rank=0 | step=18240/100000 progress=18.2% loss=0.128784 lr=9.528350e-06 elapsed=06:43:19 avg_step=1.33s recent_step=1.44s eta=30:07:55 +2026-08-25 10:39:05 | INFO | train rank=0 | step=18250/100000 progress=18.2% loss=0.062775 lr=9.527649e-06 elapsed=06:43:35 avg_step=1.33s recent_step=1.57s eta=30:07:52 +2026-08-25 10:39:18 | INFO | train rank=0 | step=18260/100000 progress=18.3% loss=0.034660 lr=9.526947e-06 elapsed=06:43:47 avg_step=1.33s recent_step=1.23s eta=30:07:35 +2026-08-25 10:39:31 | INFO | train rank=0 | step=18270/100000 progress=18.3% loss=0.046066 lr=9.526245e-06 elapsed=06:44:01 avg_step=1.33s recent_step=1.37s eta=30:07:24 +2026-08-25 10:39:47 | INFO | train rank=0 | step=18280/100000 progress=18.3% loss=0.094727 lr=9.525542e-06 elapsed=06:44:17 avg_step=1.33s recent_step=1.57s eta=30:07:21 +2026-08-25 10:40:00 | INFO | train rank=0 | step=18290/100000 progress=18.3% loss=0.059494 lr=9.524839e-06 elapsed=06:44:30 avg_step=1.33s recent_step=1.28s eta=30:07:06 +2026-08-25 10:40:12 | INFO | train rank=0 | step=18300/100000 progress=18.3% loss=0.113464 lr=9.524135e-06 elapsed=06:44:41 avg_step=1.33s recent_step=1.17s eta=30:06:46 +2026-08-25 10:40:23 | INFO | train rank=0 | step=18310/100000 progress=18.3% loss=0.070831 lr=9.523431e-06 elapsed=06:44:53 avg_step=1.33s recent_step=1.16s eta=30:06:25 +2026-08-25 10:40:41 | INFO | train rank=0 | step=18320/100000 progress=18.3% loss=0.049423 lr=9.522726e-06 elapsed=06:45:11 avg_step=1.33s recent_step=1.79s eta=30:06:32 +2026-08-25 10:40:55 | INFO | train rank=0 | step=18330/100000 progress=18.3% loss=0.067505 lr=9.522021e-06 elapsed=06:45:25 avg_step=1.33s recent_step=1.40s eta=30:06:22 +2026-08-25 10:41:08 | INFO | train rank=0 | step=18340/100000 progress=18.3% loss=0.038101 lr=9.521315e-06 elapsed=06:45:38 avg_step=1.33s recent_step=1.31s eta=30:06:08 +2026-08-25 10:41:20 | INFO | train rank=0 | step=18350/100000 progress=18.4% loss=0.052811 lr=9.520609e-06 elapsed=06:45:50 avg_step=1.33s recent_step=1.22s eta=30:05:50 +2026-08-25 10:41:33 | INFO | train rank=0 | step=18360/100000 progress=18.4% loss=0.080780 lr=9.519902e-06 elapsed=06:46:03 avg_step=1.33s recent_step=1.28s eta=30:05:35 +2026-08-25 10:41:45 | INFO | train rank=0 | step=18370/100000 progress=18.4% loss=0.076874 lr=9.519195e-06 elapsed=06:46:15 avg_step=1.33s recent_step=1.18s eta=30:05:15 +2026-08-25 10:42:00 | INFO | train rank=0 | step=18380/100000 progress=18.4% loss=0.056000 lr=9.518487e-06 elapsed=06:46:29 avg_step=1.33s recent_step=1.46s eta=30:05:08 +2026-08-25 10:42:11 | INFO | train rank=0 | step=18390/100000 progress=18.4% loss=0.101990 lr=9.517779e-06 elapsed=06:46:41 avg_step=1.33s recent_step=1.15s eta=30:04:47 +2026-08-25 10:42:24 | INFO | train rank=0 | step=18400/100000 progress=18.4% loss=0.043533 lr=9.517070e-06 elapsed=06:46:54 avg_step=1.33s recent_step=1.30s eta=30:04:32 +2026-08-25 10:42:38 | INFO | train rank=0 | step=18410/100000 progress=18.4% loss=0.059143 lr=9.516361e-06 elapsed=06:47:08 avg_step=1.33s recent_step=1.43s eta=30:04:24 +2026-08-25 10:42:52 | INFO | train rank=0 | step=18420/100000 progress=18.4% loss=0.039413 lr=9.515652e-06 elapsed=06:47:22 avg_step=1.33s recent_step=1.35s eta=30:04:11 +2026-08-25 10:43:04 | INFO | train rank=0 | step=18430/100000 progress=18.4% loss=0.047272 lr=9.514941e-06 elapsed=06:47:34 avg_step=1.33s recent_step=1.18s eta=30:03:52 +2026-08-25 10:43:15 | INFO | train rank=0 | step=18440/100000 progress=18.4% loss=0.067139 lr=9.514231e-06 elapsed=06:47:45 avg_step=1.33s recent_step=1.15s eta=30:03:31 +2026-08-25 10:43:27 | INFO | train rank=0 | step=18450/100000 progress=18.4% loss=0.067902 lr=9.513519e-06 elapsed=06:47:57 avg_step=1.33s recent_step=1.20s eta=30:03:12 +2026-08-25 10:43:39 | INFO | train rank=0 | step=18460/100000 progress=18.5% loss=0.056335 lr=9.512808e-06 elapsed=06:48:08 avg_step=1.33s recent_step=1.14s eta=30:02:50 +2026-08-25 10:43:50 | INFO | train rank=0 | step=18470/100000 progress=18.5% loss=0.071228 lr=9.512096e-06 elapsed=06:48:20 avg_step=1.33s recent_step=1.17s eta=30:02:30 +2026-08-25 10:44:07 | INFO | train rank=0 | step=18480/100000 progress=18.5% loss=0.085205 lr=9.511383e-06 elapsed=06:48:37 avg_step=1.33s recent_step=1.64s eta=30:02:31 +2026-08-25 10:44:22 | INFO | train rank=0 | step=18490/100000 progress=18.5% loss=0.063736 lr=9.510670e-06 elapsed=06:48:52 avg_step=1.33s recent_step=1.51s eta=30:02:26 +2026-08-25 10:44:37 | INFO | train rank=0 | step=18500/100000 progress=18.5% loss=0.051971 lr=9.509956e-06 elapsed=06:49:07 avg_step=1.33s recent_step=1.48s eta=30:02:19 +2026-08-25 10:44:49 | INFO | train rank=0 | step=18510/100000 progress=18.5% loss=0.034081 lr=9.509242e-06 elapsed=06:49:19 avg_step=1.33s recent_step=1.21s eta=30:02:01 +2026-08-25 10:45:03 | INFO | train rank=0 | step=18520/100000 progress=18.5% loss=0.068665 lr=9.508527e-06 elapsed=06:49:33 avg_step=1.33s recent_step=1.39s eta=30:01:50 +2026-08-25 10:45:15 | INFO | train rank=0 | step=18530/100000 progress=18.5% loss=0.074158 lr=9.507812e-06 elapsed=06:49:45 avg_step=1.33s recent_step=1.23s eta=30:01:32 +2026-08-25 10:45:27 | INFO | train rank=0 | step=18540/100000 progress=18.5% loss=0.057129 lr=9.507097e-06 elapsed=06:49:57 avg_step=1.33s recent_step=1.18s eta=30:01:13 +2026-08-25 10:45:39 | INFO | train rank=0 | step=18550/100000 progress=18.6% loss=0.044479 lr=9.506381e-06 elapsed=06:50:09 avg_step=1.33s recent_step=1.27s eta=30:00:57 +2026-08-25 10:45:55 | INFO | train rank=0 | step=18560/100000 progress=18.6% loss=0.039322 lr=9.505664e-06 elapsed=06:50:25 avg_step=1.33s recent_step=1.55s eta=30:00:53 +2026-08-25 10:46:07 | INFO | train rank=0 | step=18570/100000 progress=18.6% loss=0.066544 lr=9.504947e-06 elapsed=06:50:37 avg_step=1.33s recent_step=1.26s eta=30:00:37 +2026-08-25 10:46:19 | INFO | train rank=0 | step=18580/100000 progress=18.6% loss=0.045898 lr=9.504229e-06 elapsed=06:50:49 avg_step=1.33s recent_step=1.18s eta=30:00:18 +2026-08-25 10:46:32 | INFO | train rank=0 | step=18590/100000 progress=18.6% loss=0.059753 lr=9.503511e-06 elapsed=06:51:02 avg_step=1.33s recent_step=1.30s eta=30:00:03 +2026-08-25 10:46:45 | INFO | train rank=0 | step=18600/100000 progress=18.6% loss=0.066406 lr=9.502793e-06 elapsed=06:51:15 avg_step=1.33s recent_step=1.27s eta=29:59:47 +2026-08-25 10:47:00 | INFO | train rank=0 | step=18610/100000 progress=18.6% loss=0.081635 lr=9.502074e-06 elapsed=06:51:30 avg_step=1.33s recent_step=1.52s eta=29:59:43 +2026-08-25 10:47:12 | INFO | train rank=0 | step=18620/100000 progress=18.6% loss=0.047287 lr=9.501354e-06 elapsed=06:51:42 avg_step=1.33s recent_step=1.17s eta=29:59:22 +2026-08-25 10:47:30 | INFO | train rank=0 | step=18630/100000 progress=18.6% loss=0.055527 lr=9.500634e-06 elapsed=06:52:00 avg_step=1.33s recent_step=1.82s eta=29:59:31 +2026-08-25 10:47:42 | INFO | train rank=0 | step=18640/100000 progress=18.6% loss=0.058319 lr=9.499913e-06 elapsed=06:52:12 avg_step=1.33s recent_step=1.21s eta=29:59:12 +2026-08-25 10:47:55 | INFO | train rank=0 | step=18650/100000 progress=18.6% loss=0.059998 lr=9.499192e-06 elapsed=06:52:25 avg_step=1.33s recent_step=1.25s eta=29:58:56 +2026-08-25 10:48:06 | INFO | train rank=0 | step=18660/100000 progress=18.7% loss=0.047638 lr=9.498471e-06 elapsed=06:52:36 avg_step=1.33s recent_step=1.13s eta=29:58:34 +2026-08-25 10:48:20 | INFO | train rank=0 | step=18670/100000 progress=18.7% loss=0.048615 lr=9.497749e-06 elapsed=06:52:50 avg_step=1.33s recent_step=1.45s eta=29:58:26 +2026-08-25 10:48:32 | INFO | train rank=0 | step=18680/100000 progress=18.7% loss=0.082825 lr=9.497026e-06 elapsed=06:53:01 avg_step=1.33s recent_step=1.12s eta=29:58:03 +2026-08-25 10:48:46 | INFO | train rank=0 | step=18690/100000 progress=18.7% loss=0.136597 lr=9.496303e-06 elapsed=06:53:16 avg_step=1.33s recent_step=1.41s eta=29:57:54 +2026-08-25 10:49:02 | INFO | train rank=0 | step=18700/100000 progress=18.7% loss=0.134155 lr=9.495580e-06 elapsed=06:53:32 avg_step=1.33s recent_step=1.60s eta=29:57:52 +2026-08-25 10:49:16 | INFO | train rank=0 | step=18710/100000 progress=18.7% loss=0.075500 lr=9.494856e-06 elapsed=06:53:46 avg_step=1.33s recent_step=1.40s eta=29:57:42 +2026-08-25 10:49:33 | INFO | train rank=0 | step=18720/100000 progress=18.7% loss=0.115112 lr=9.494131e-06 elapsed=06:54:03 avg_step=1.33s recent_step=1.74s eta=29:57:47 +2026-08-25 10:49:44 | INFO | train rank=0 | step=18730/100000 progress=18.7% loss=0.072083 lr=9.493406e-06 elapsed=06:54:14 avg_step=1.33s recent_step=1.14s eta=29:57:25 +2026-08-25 10:49:56 | INFO | train rank=0 | step=18740/100000 progress=18.7% loss=0.059875 lr=9.492681e-06 elapsed=06:54:26 avg_step=1.33s recent_step=1.18s eta=29:57:05 +2026-08-25 10:50:08 | INFO | train rank=0 | step=18750/100000 progress=18.8% loss=0.063324 lr=9.491955e-06 elapsed=06:54:38 avg_step=1.33s recent_step=1.15s eta=29:56:45 +2026-08-25 10:50:21 | INFO | train rank=0 | step=18760/100000 progress=18.8% loss=0.046097 lr=9.491228e-06 elapsed=06:54:51 avg_step=1.33s recent_step=1.35s eta=29:56:32 +2026-08-25 10:50:33 | INFO | train rank=0 | step=18770/100000 progress=18.8% loss=0.064850 lr=9.490502e-06 elapsed=06:55:03 avg_step=1.33s recent_step=1.19s eta=29:56:13 +2026-08-25 10:50:44 | INFO | train rank=0 | step=18780/100000 progress=18.8% loss=0.055969 lr=9.489774e-06 elapsed=06:55:14 avg_step=1.33s recent_step=1.12s eta=29:55:51 +2026-08-25 10:50:56 | INFO | train rank=0 | step=18790/100000 progress=18.8% loss=0.082275 lr=9.489046e-06 elapsed=06:55:26 avg_step=1.33s recent_step=1.15s eta=29:55:30 +2026-08-25 10:51:07 | INFO | train rank=0 | step=18800/100000 progress=18.8% loss=0.069504 lr=9.488318e-06 elapsed=06:55:37 avg_step=1.33s recent_step=1.16s eta=29:55:09 +2026-08-25 10:51:21 | INFO | train rank=0 | step=18810/100000 progress=18.8% loss=0.053940 lr=9.487589e-06 elapsed=06:55:51 avg_step=1.33s recent_step=1.39s eta=29:54:59 +2026-08-25 10:51:34 | INFO | train rank=0 | step=18820/100000 progress=18.8% loss=0.062714 lr=9.486859e-06 elapsed=06:56:04 avg_step=1.33s recent_step=1.29s eta=29:54:44 +2026-08-25 10:51:49 | INFO | train rank=0 | step=18830/100000 progress=18.8% loss=0.052399 lr=9.486130e-06 elapsed=06:56:19 avg_step=1.33s recent_step=1.53s eta=29:54:39 +2026-08-25 10:52:01 | INFO | train rank=0 | step=18840/100000 progress=18.8% loss=0.092224 lr=9.485399e-06 elapsed=06:56:31 avg_step=1.33s recent_step=1.20s eta=29:54:21 +2026-08-25 10:52:16 | INFO | train rank=0 | step=18850/100000 progress=18.9% loss=0.086975 lr=9.484668e-06 elapsed=06:56:46 avg_step=1.33s recent_step=1.43s eta=29:54:12 +2026-08-25 10:52:28 | INFO | train rank=0 | step=18860/100000 progress=18.9% loss=0.066010 lr=9.483937e-06 elapsed=06:56:58 avg_step=1.33s recent_step=1.25s eta=29:53:55 +2026-08-25 10:52:40 | INFO | train rank=0 | step=18870/100000 progress=18.9% loss=0.096436 lr=9.483205e-06 elapsed=06:57:10 avg_step=1.33s recent_step=1.20s eta=29:53:36 +2026-08-25 10:52:52 | INFO | train rank=0 | step=18880/100000 progress=18.9% loss=0.066940 lr=9.482473e-06 elapsed=06:57:22 avg_step=1.33s recent_step=1.15s eta=29:53:16 +2026-08-25 10:53:03 | INFO | train rank=0 | step=18890/100000 progress=18.9% loss=0.039032 lr=9.481740e-06 elapsed=06:57:33 avg_step=1.33s recent_step=1.13s eta=29:52:54 +2026-08-25 10:53:15 | INFO | train rank=0 | step=18900/100000 progress=18.9% loss=0.073486 lr=9.481007e-06 elapsed=06:57:45 avg_step=1.33s recent_step=1.22s eta=29:52:36 +2026-08-25 10:53:27 | INFO | train rank=0 | step=18910/100000 progress=18.9% loss=0.078735 lr=9.480273e-06 elapsed=06:57:57 avg_step=1.33s recent_step=1.17s eta=29:52:16 +2026-08-25 10:53:41 | INFO | train rank=0 | step=18920/100000 progress=18.9% loss=0.078491 lr=9.479539e-06 elapsed=06:58:11 avg_step=1.33s recent_step=1.40s eta=29:52:06 +2026-08-25 10:53:56 | INFO | train rank=0 | step=18930/100000 progress=18.9% loss=0.053009 lr=9.478804e-06 elapsed=06:58:26 avg_step=1.33s recent_step=1.54s eta=29:52:01 +2026-08-25 10:54:09 | INFO | train rank=0 | step=18940/100000 progress=18.9% loss=0.046463 lr=9.478069e-06 elapsed=06:58:39 avg_step=1.33s recent_step=1.26s eta=29:51:45 +2026-08-25 10:54:24 | INFO | train rank=0 | step=18950/100000 progress=18.9% loss=0.109619 lr=9.477333e-06 elapsed=06:58:54 avg_step=1.33s recent_step=1.50s eta=29:51:39 +2026-08-25 10:54:40 | INFO | train rank=0 | step=18960/100000 progress=19.0% loss=0.068436 lr=9.476597e-06 elapsed=06:59:10 avg_step=1.33s recent_step=1.60s eta=29:51:38 +2026-08-25 10:54:53 | INFO | train rank=0 | step=18970/100000 progress=19.0% loss=0.038162 lr=9.475860e-06 elapsed=06:59:23 avg_step=1.33s recent_step=1.29s eta=29:51:23 +2026-08-25 10:55:07 | INFO | train rank=0 | step=18980/100000 progress=19.0% loss=0.074219 lr=9.475123e-06 elapsed=06:59:37 avg_step=1.33s recent_step=1.49s eta=29:51:16 +2026-08-25 10:55:22 | INFO | train rank=0 | step=18990/100000 progress=19.0% loss=0.056976 lr=9.474385e-06 elapsed=06:59:52 avg_step=1.33s recent_step=1.50s eta=29:51:10 +2026-08-25 10:55:37 | INFO | train rank=0 | step=19000/100000 progress=19.0% loss=0.053436 lr=9.473647e-06 elapsed=07:00:07 avg_step=1.33s recent_step=1.49s eta=29:51:04 +2026-08-25 10:55:49 | INFO | train rank=0 | step=19010/100000 progress=19.0% loss=0.048019 lr=9.472908e-06 elapsed=07:00:19 avg_step=1.33s recent_step=1.17s eta=29:50:44 +2026-08-25 10:56:04 | INFO | train rank=0 | step=19020/100000 progress=19.0% loss=0.049622 lr=9.472169e-06 elapsed=07:00:34 avg_step=1.33s recent_step=1.46s eta=29:50:37 +2026-08-25 10:56:15 | INFO | train rank=0 | step=19030/100000 progress=19.0% loss=0.072723 lr=9.471429e-06 elapsed=07:00:45 avg_step=1.33s recent_step=1.15s eta=29:50:16 +2026-08-25 10:56:28 | INFO | train rank=0 | step=19040/100000 progress=19.0% loss=0.147522 lr=9.470689e-06 elapsed=07:00:58 avg_step=1.33s recent_step=1.30s eta=29:50:02 +2026-08-25 10:56:40 | INFO | train rank=0 | step=19050/100000 progress=19.1% loss=0.070068 lr=9.469948e-06 elapsed=07:01:10 avg_step=1.33s recent_step=1.14s eta=29:49:41 +2026-08-25 10:56:54 | INFO | train rank=0 | step=19060/100000 progress=19.1% loss=0.067322 lr=9.469207e-06 elapsed=07:01:24 avg_step=1.33s recent_step=1.40s eta=29:49:31 +2026-08-25 10:57:07 | INFO | train rank=0 | step=19070/100000 progress=19.1% loss=0.049622 lr=9.468465e-06 elapsed=07:01:37 avg_step=1.33s recent_step=1.29s eta=29:49:16 +2026-08-25 10:57:20 | INFO | train rank=0 | step=19080/100000 progress=19.1% loss=0.050385 lr=9.467723e-06 elapsed=07:01:50 avg_step=1.33s recent_step=1.33s eta=29:49:03 +2026-08-25 10:57:33 | INFO | train rank=0 | step=19090/100000 progress=19.1% loss=0.051285 lr=9.466981e-06 elapsed=07:02:03 avg_step=1.33s recent_step=1.29s eta=29:48:48 +2026-08-25 10:57:45 | INFO | train rank=0 | step=19100/100000 progress=19.1% loss=0.071533 lr=9.466238e-06 elapsed=07:02:15 avg_step=1.33s recent_step=1.24s eta=29:48:31 +2026-08-25 10:58:00 | INFO | train rank=0 | step=19110/100000 progress=19.1% loss=0.084442 lr=9.465494e-06 elapsed=07:02:30 avg_step=1.33s recent_step=1.50s eta=29:48:25 +2026-08-25 10:58:15 | INFO | train rank=0 | step=19120/100000 progress=19.1% loss=0.112549 lr=9.464750e-06 elapsed=07:02:45 avg_step=1.33s recent_step=1.47s eta=29:48:18 +2026-08-25 10:58:29 | INFO | train rank=0 | step=19130/100000 progress=19.1% loss=0.054352 lr=9.464005e-06 elapsed=07:02:59 avg_step=1.33s recent_step=1.37s eta=29:48:07 +2026-08-25 10:58:41 | INFO | train rank=0 | step=19140/100000 progress=19.1% loss=0.037537 lr=9.463260e-06 elapsed=07:03:11 avg_step=1.33s recent_step=1.28s eta=29:47:51 +2026-08-25 10:58:54 | INFO | train rank=0 | step=19150/100000 progress=19.1% loss=0.042725 lr=9.462515e-06 elapsed=07:03:24 avg_step=1.33s recent_step=1.23s eta=29:47:34 +2026-08-25 10:59:09 | INFO | train rank=0 | step=19160/100000 progress=19.2% loss=0.054626 lr=9.461769e-06 elapsed=07:03:39 avg_step=1.33s recent_step=1.53s eta=29:47:29 +2026-08-25 10:59:21 | INFO | train rank=0 | step=19170/100000 progress=19.2% loss=0.064606 lr=9.461022e-06 elapsed=07:03:51 avg_step=1.33s recent_step=1.20s eta=29:47:10 +2026-08-25 10:59:34 | INFO | train rank=0 | step=19180/100000 progress=19.2% loss=0.068939 lr=9.460275e-06 elapsed=07:04:04 avg_step=1.33s recent_step=1.32s eta=29:46:57 +2026-08-25 10:59:51 | INFO | train rank=0 | step=19190/100000 progress=19.2% loss=0.060791 lr=9.459528e-06 elapsed=07:04:20 avg_step=1.33s recent_step=1.64s eta=29:46:57 +2026-08-25 11:00:06 | INFO | train rank=0 | step=19200/100000 progress=19.2% loss=0.066803 lr=9.458780e-06 elapsed=07:04:35 avg_step=1.33s recent_step=1.50s eta=29:46:51 +2026-08-25 11:00:17 | INFO | train rank=0 | step=19210/100000 progress=19.2% loss=0.056641 lr=9.458031e-06 elapsed=07:04:47 avg_step=1.33s recent_step=1.17s eta=29:46:31 +2026-08-25 11:00:30 | INFO | train rank=0 | step=19220/100000 progress=19.2% loss=0.084137 lr=9.457282e-06 elapsed=07:05:00 avg_step=1.33s recent_step=1.26s eta=29:46:15 +2026-08-25 11:00:45 | INFO | train rank=0 | step=19230/100000 progress=19.2% loss=0.051910 lr=9.456533e-06 elapsed=07:05:14 avg_step=1.33s recent_step=1.47s eta=29:46:08 +2026-08-25 11:00:56 | INFO | train rank=0 | step=19240/100000 progress=19.2% loss=0.069611 lr=9.455783e-06 elapsed=07:05:26 avg_step=1.33s recent_step=1.19s eta=29:45:49 +2026-08-25 11:01:14 | INFO | train rank=0 | step=19250/100000 progress=19.2% loss=0.061249 lr=9.455033e-06 elapsed=07:05:44 avg_step=1.33s recent_step=1.73s eta=29:45:52 +2026-08-25 11:01:25 | INFO | train rank=0 | step=19260/100000 progress=19.3% loss=0.027557 lr=9.454282e-06 elapsed=07:05:55 avg_step=1.33s recent_step=1.09s eta=29:45:29 +2026-08-25 11:01:43 | INFO | train rank=0 | step=19270/100000 progress=19.3% loss=0.101074 lr=9.453530e-06 elapsed=07:06:13 avg_step=1.33s recent_step=1.82s eta=29:45:36 +2026-08-25 11:01:58 | INFO | train rank=0 | step=19280/100000 progress=19.3% loss=0.057861 lr=9.452778e-06 elapsed=07:06:28 avg_step=1.33s recent_step=1.51s eta=29:45:31 +2026-08-25 11:02:09 | INFO | train rank=0 | step=19290/100000 progress=19.3% loss=0.053223 lr=9.452026e-06 elapsed=07:06:39 avg_step=1.33s recent_step=1.14s eta=29:45:10 +2026-08-25 11:02:21 | INFO | train rank=0 | step=19300/100000 progress=19.3% loss=0.058487 lr=9.451273e-06 elapsed=07:06:51 avg_step=1.33s recent_step=1.17s eta=29:44:50 +2026-08-25 11:02:35 | INFO | train rank=0 | step=19310/100000 progress=19.3% loss=0.052124 lr=9.450520e-06 elapsed=07:07:05 avg_step=1.33s recent_step=1.41s eta=29:44:40 +2026-08-25 11:02:47 | INFO | train rank=0 | step=19320/100000 progress=19.3% loss=0.062469 lr=9.449766e-06 elapsed=07:07:17 avg_step=1.33s recent_step=1.17s eta=29:44:21 +2026-08-25 11:02:59 | INFO | train rank=0 | step=19330/100000 progress=19.3% loss=0.073029 lr=9.449012e-06 elapsed=07:07:29 avg_step=1.33s recent_step=1.17s eta=29:44:01 +2026-08-25 11:03:10 | INFO | train rank=0 | step=19340/100000 progress=19.3% loss=0.053009 lr=9.448257e-06 elapsed=07:07:40 avg_step=1.33s recent_step=1.13s eta=29:43:39 +2026-08-25 11:03:26 | INFO | train rank=0 | step=19350/100000 progress=19.4% loss=0.064331 lr=9.447502e-06 elapsed=07:07:56 avg_step=1.33s recent_step=1.64s eta=29:43:39 +2026-08-25 11:03:37 | INFO | train rank=0 | step=19360/100000 progress=19.4% loss=0.058075 lr=9.446746e-06 elapsed=07:08:07 avg_step=1.33s recent_step=1.10s eta=29:43:17 +2026-08-25 11:03:49 | INFO | train rank=0 | step=19370/100000 progress=19.4% loss=0.097443 lr=9.445990e-06 elapsed=07:08:19 avg_step=1.33s recent_step=1.14s eta=29:42:55 +2026-08-25 11:04:00 | INFO | train rank=0 | step=19380/100000 progress=19.4% loss=0.052979 lr=9.445233e-06 elapsed=07:08:30 avg_step=1.33s recent_step=1.17s eta=29:42:36 +2026-08-25 11:04:13 | INFO | train rank=0 | step=19390/100000 progress=19.4% loss=0.057098 lr=9.444476e-06 elapsed=07:08:43 avg_step=1.33s recent_step=1.26s eta=29:42:19 +2026-08-25 11:04:31 | INFO | train rank=0 | step=19400/100000 progress=19.4% loss=0.057449 lr=9.443718e-06 elapsed=07:09:01 avg_step=1.33s recent_step=1.81s eta=29:42:26 +2026-08-25 11:04:44 | INFO | train rank=0 | step=19410/100000 progress=19.4% loss=0.050003 lr=9.442960e-06 elapsed=07:09:14 avg_step=1.33s recent_step=1.25s eta=29:42:10 +2026-08-25 11:04:58 | INFO | train rank=0 | step=19420/100000 progress=19.4% loss=0.068039 lr=9.442201e-06 elapsed=07:09:28 avg_step=1.33s recent_step=1.42s eta=29:42:00 +2026-08-25 11:05:10 | INFO | train rank=0 | step=19430/100000 progress=19.4% loss=0.076904 lr=9.441442e-06 elapsed=07:09:40 avg_step=1.33s recent_step=1.25s eta=29:41:44 +2026-08-25 11:05:24 | INFO | train rank=0 | step=19440/100000 progress=19.4% loss=0.080994 lr=9.440682e-06 elapsed=07:09:54 avg_step=1.33s recent_step=1.33s eta=29:41:31 +2026-08-25 11:05:35 | INFO | train rank=0 | step=19450/100000 progress=19.4% loss=0.080688 lr=9.439922e-06 elapsed=07:10:05 avg_step=1.33s recent_step=1.14s eta=29:41:10 +2026-08-25 11:05:47 | INFO | train rank=0 | step=19460/100000 progress=19.5% loss=0.089996 lr=9.439161e-06 elapsed=07:10:17 avg_step=1.33s recent_step=1.18s eta=29:40:50 +2026-08-25 11:06:06 | INFO | train rank=0 | step=19470/100000 progress=19.5% loss=0.078583 lr=9.438400e-06 elapsed=07:10:35 avg_step=1.33s recent_step=1.88s eta=29:41:00 +2026-08-25 11:06:18 | INFO | train rank=0 | step=19480/100000 progress=19.5% loss=0.051575 lr=9.437639e-06 elapsed=07:10:47 avg_step=1.33s recent_step=1.20s eta=29:40:41 +2026-08-25 11:06:34 | INFO | train rank=0 | step=19490/100000 progress=19.5% loss=0.047241 lr=9.436877e-06 elapsed=07:11:04 avg_step=1.33s recent_step=1.66s eta=29:40:42 +2026-08-25 11:06:45 | INFO | train rank=0 | step=19500/100000 progress=19.5% loss=0.084076 lr=9.436114e-06 elapsed=07:11:15 avg_step=1.33s recent_step=1.10s eta=29:40:19 +2026-08-25 11:06:59 | INFO | train rank=0 | step=19510/100000 progress=19.5% loss=0.082794 lr=9.435351e-06 elapsed=07:11:29 avg_step=1.33s recent_step=1.41s eta=29:40:09 +2026-08-25 11:07:12 | INFO | train rank=0 | step=19520/100000 progress=19.5% loss=0.073502 lr=9.434587e-06 elapsed=07:11:42 avg_step=1.33s recent_step=1.32s eta=29:39:56 +2026-08-25 11:07:30 | INFO | train rank=0 | step=19530/100000 progress=19.5% loss=0.065872 lr=9.433823e-06 elapsed=07:12:00 avg_step=1.33s recent_step=1.77s eta=29:40:01 +2026-08-25 11:07:42 | INFO | train rank=0 | step=19540/100000 progress=19.5% loss=0.056847 lr=9.433059e-06 elapsed=07:12:12 avg_step=1.33s recent_step=1.21s eta=29:39:43 +2026-08-25 11:07:55 | INFO | train rank=0 | step=19550/100000 progress=19.6% loss=0.060760 lr=9.432294e-06 elapsed=07:12:25 avg_step=1.33s recent_step=1.25s eta=29:39:26 +2026-08-25 11:08:06 | INFO | train rank=0 | step=19560/100000 progress=19.6% loss=0.044136 lr=9.431528e-06 elapsed=07:12:36 avg_step=1.33s recent_step=1.16s eta=29:39:06 +2026-08-25 11:08:24 | INFO | train rank=0 | step=19570/100000 progress=19.6% loss=0.080170 lr=9.430763e-06 elapsed=07:12:54 avg_step=1.33s recent_step=1.78s eta=29:39:12 +2026-08-25 11:08:36 | INFO | train rank=0 | step=19580/100000 progress=19.6% loss=0.072693 lr=9.429996e-06 elapsed=07:13:05 avg_step=1.33s recent_step=1.12s eta=29:38:50 +2026-08-25 11:08:53 | INFO | train rank=0 | step=19590/100000 progress=19.6% loss=0.062408 lr=9.429229e-06 elapsed=07:13:23 avg_step=1.33s recent_step=1.75s eta=29:38:54 +2026-08-25 11:09:05 | INFO | train rank=0 | step=19600/100000 progress=19.6% loss=0.067276 lr=9.428462e-06 elapsed=07:13:34 avg_step=1.33s recent_step=1.15s eta=29:38:34 +2026-08-25 11:09:17 | INFO | train rank=0 | step=19610/100000 progress=19.6% loss=0.064148 lr=9.427694e-06 elapsed=07:13:47 avg_step=1.33s recent_step=1.29s eta=29:38:19 +2026-08-25 11:09:29 | INFO | train rank=0 | step=19620/100000 progress=19.6% loss=0.038895 lr=9.426925e-06 elapsed=07:13:59 avg_step=1.33s recent_step=1.19s eta=29:38:00 +2026-08-25 11:09:48 | INFO | train rank=0 | step=19630/100000 progress=19.6% loss=0.050659 lr=9.426157e-06 elapsed=07:14:18 avg_step=1.33s recent_step=1.88s eta=29:38:09 +2026-08-25 11:10:00 | INFO | train rank=0 | step=19640/100000 progress=19.6% loss=0.097046 lr=9.425387e-06 elapsed=07:14:30 avg_step=1.33s recent_step=1.20s eta=29:37:51 +2026-08-25 11:10:12 | INFO | train rank=0 | step=19650/100000 progress=19.7% loss=0.102783 lr=9.424617e-06 elapsed=07:14:42 avg_step=1.33s recent_step=1.16s eta=29:37:31 +2026-08-25 11:10:23 | INFO | train rank=0 | step=19660/100000 progress=19.7% loss=0.065491 lr=9.423847e-06 elapsed=07:14:53 avg_step=1.33s recent_step=1.15s eta=29:37:10 +2026-08-25 11:10:41 | INFO | train rank=0 | step=19670/100000 progress=19.7% loss=0.054321 lr=9.423076e-06 elapsed=07:15:11 avg_step=1.33s recent_step=1.79s eta=29:37:16 +2026-08-25 11:10:54 | INFO | train rank=0 | step=19680/100000 progress=19.7% loss=0.053833 lr=9.422305e-06 elapsed=07:15:24 avg_step=1.33s recent_step=1.33s eta=29:37:02 +2026-08-25 11:11:07 | INFO | train rank=0 | step=19690/100000 progress=19.7% loss=0.107117 lr=9.421533e-06 elapsed=07:15:37 avg_step=1.33s recent_step=1.28s eta=29:36:47 +2026-08-25 11:11:19 | INFO | train rank=0 | step=19700/100000 progress=19.7% loss=0.047455 lr=9.420761e-06 elapsed=07:15:49 avg_step=1.33s recent_step=1.21s eta=29:36:29 +2026-08-25 11:11:30 | INFO | train rank=0 | step=19710/100000 progress=19.7% loss=0.061874 lr=9.419988e-06 elapsed=07:16:00 avg_step=1.33s recent_step=1.12s eta=29:36:07 +2026-08-25 11:11:44 | INFO | train rank=0 | step=19720/100000 progress=19.7% loss=0.054199 lr=9.419215e-06 elapsed=07:16:14 avg_step=1.33s recent_step=1.37s eta=29:35:56 +2026-08-25 11:12:00 | INFO | train rank=0 | step=19730/100000 progress=19.7% loss=0.091553 lr=9.418441e-06 elapsed=07:16:30 avg_step=1.33s recent_step=1.57s eta=29:35:52 +2026-08-25 11:12:11 | INFO | train rank=0 | step=19740/100000 progress=19.7% loss=0.066010 lr=9.417667e-06 elapsed=07:16:41 avg_step=1.33s recent_step=1.15s eta=29:35:32 +2026-08-25 11:12:24 | INFO | train rank=0 | step=19750/100000 progress=19.8% loss=0.071564 lr=9.416892e-06 elapsed=07:16:54 avg_step=1.33s recent_step=1.27s eta=29:35:16 +2026-08-25 11:12:39 | INFO | train rank=0 | step=19760/100000 progress=19.8% loss=0.061249 lr=9.416117e-06 elapsed=07:17:09 avg_step=1.33s recent_step=1.46s eta=29:35:08 +2026-08-25 11:12:50 | INFO | train rank=0 | step=19770/100000 progress=19.8% loss=0.054779 lr=9.415342e-06 elapsed=07:17:20 avg_step=1.33s recent_step=1.15s eta=29:34:48 +2026-08-25 11:13:04 | INFO | train rank=0 | step=19780/100000 progress=19.8% loss=0.061371 lr=9.414565e-06 elapsed=07:17:34 avg_step=1.33s recent_step=1.36s eta=29:34:36 +2026-08-25 11:13:22 | INFO | train rank=0 | step=19790/100000 progress=19.8% loss=0.067307 lr=9.413789e-06 elapsed=07:17:52 avg_step=1.33s recent_step=1.87s eta=29:34:45 +2026-08-25 11:13:34 | INFO | train rank=0 | step=19800/100000 progress=19.8% loss=0.045242 lr=9.413012e-06 elapsed=07:18:04 avg_step=1.33s recent_step=1.14s eta=29:34:24 +2026-08-25 11:13:45 | INFO | train rank=0 | step=19810/100000 progress=19.8% loss=0.058929 lr=9.412234e-06 elapsed=07:18:15 avg_step=1.33s recent_step=1.12s eta=29:34:02 +2026-08-25 11:13:57 | INFO | train rank=0 | step=19820/100000 progress=19.8% loss=0.047546 lr=9.411456e-06 elapsed=07:18:27 avg_step=1.33s recent_step=1.19s eta=29:33:43 +2026-08-25 11:14:10 | INFO | train rank=0 | step=19830/100000 progress=19.8% loss=0.072723 lr=9.410678e-06 elapsed=07:18:40 avg_step=1.33s recent_step=1.33s eta=29:33:30 +2026-08-25 11:14:22 | INFO | train rank=0 | step=19840/100000 progress=19.8% loss=0.039291 lr=9.409899e-06 elapsed=07:18:51 avg_step=1.33s recent_step=1.13s eta=29:33:09 +2026-08-25 11:14:33 | INFO | train rank=0 | step=19850/100000 progress=19.9% loss=0.077057 lr=9.409119e-06 elapsed=07:19:03 avg_step=1.33s recent_step=1.19s eta=29:32:50 +2026-08-25 11:14:51 | INFO | train rank=0 | step=19860/100000 progress=19.9% loss=0.052414 lr=9.408339e-06 elapsed=07:19:21 avg_step=1.33s recent_step=1.77s eta=29:32:55 +2026-08-25 11:15:05 | INFO | train rank=0 | step=19870/100000 progress=19.9% loss=0.073334 lr=9.407559e-06 elapsed=07:19:35 avg_step=1.33s recent_step=1.37s eta=29:32:43 +2026-08-25 11:15:16 | INFO | train rank=0 | step=19880/100000 progress=19.9% loss=0.079773 lr=9.406778e-06 elapsed=07:19:46 avg_step=1.33s recent_step=1.17s eta=29:32:23 +2026-08-25 11:15:28 | INFO | train rank=0 | step=19890/100000 progress=19.9% loss=0.036041 lr=9.405996e-06 elapsed=07:19:58 avg_step=1.33s recent_step=1.17s eta=29:32:04 +2026-08-25 11:15:42 | INFO | train rank=0 | step=19900/100000 progress=19.9% loss=0.063751 lr=9.405214e-06 elapsed=07:20:12 avg_step=1.33s recent_step=1.36s eta=29:31:52 +2026-08-25 11:15:56 | INFO | train rank=0 | step=19910/100000 progress=19.9% loss=0.052704 lr=9.404432e-06 elapsed=07:20:26 avg_step=1.33s recent_step=1.46s eta=29:31:44 +2026-08-25 11:16:08 | INFO | train rank=0 | step=19920/100000 progress=19.9% loss=0.059753 lr=9.403649e-06 elapsed=07:20:38 avg_step=1.33s recent_step=1.16s eta=29:31:24 +2026-08-25 11:16:22 | INFO | train rank=0 | step=19930/100000 progress=19.9% loss=0.073105 lr=9.402866e-06 elapsed=07:20:52 avg_step=1.33s recent_step=1.41s eta=29:31:14 +2026-08-25 11:16:33 | INFO | train rank=0 | step=19940/100000 progress=19.9% loss=0.059052 lr=9.402082e-06 elapsed=07:21:03 avg_step=1.33s recent_step=1.14s eta=29:30:53 +2026-08-25 11:16:48 | INFO | train rank=0 | step=19950/100000 progress=20.0% loss=0.064713 lr=9.401298e-06 elapsed=07:21:18 avg_step=1.33s recent_step=1.46s eta=29:30:45 +2026-08-25 11:17:02 | INFO | train rank=0 | step=19960/100000 progress=20.0% loss=0.055481 lr=9.400513e-06 elapsed=07:21:32 avg_step=1.33s recent_step=1.37s eta=29:30:34 +2026-08-25 11:17:14 | INFO | train rank=0 | step=19970/100000 progress=20.0% loss=0.085724 lr=9.399727e-06 elapsed=07:21:44 avg_step=1.33s recent_step=1.22s eta=29:30:16 +2026-08-25 11:17:26 | INFO | train rank=0 | step=19980/100000 progress=20.0% loss=0.050537 lr=9.398942e-06 elapsed=07:21:56 avg_step=1.33s recent_step=1.25s eta=29:30:00 +2026-08-25 11:17:40 | INFO | train rank=0 | step=19990/100000 progress=20.0% loss=0.061447 lr=9.398155e-06 elapsed=07:22:10 avg_step=1.33s recent_step=1.33s eta=29:29:47 +2026-08-25 11:17:51 | INFO | train rank=0 | step=20000/100000 progress=20.0% loss=0.070282 lr=9.397369e-06 elapsed=07:22:21 avg_step=1.33s recent_step=1.13s eta=29:29:25 +2026-08-25 11:18:04 | INFO | train rank=0 | step=20010/100000 progress=20.0% loss=0.100830 lr=9.396582e-06 elapsed=07:22:34 avg_step=1.33s recent_step=1.30s eta=29:29:11 +2026-08-25 11:18:19 | INFO | train rank=0 | step=20020/100000 progress=20.0% loss=0.069336 lr=9.395794e-06 elapsed=07:22:49 avg_step=1.33s recent_step=1.50s eta=29:29:04 +2026-08-25 11:18:30 | INFO | train rank=0 | step=20030/100000 progress=20.0% loss=0.092285 lr=9.395006e-06 elapsed=07:23:00 avg_step=1.33s recent_step=1.14s eta=29:28:44 +2026-08-25 11:18:42 | INFO | train rank=0 | step=20040/100000 progress=20.0% loss=0.081146 lr=9.394217e-06 elapsed=07:23:12 avg_step=1.33s recent_step=1.13s eta=29:28:23 +2026-08-25 11:18:56 | INFO | train rank=0 | step=20050/100000 progress=20.1% loss=0.033951 lr=9.393428e-06 elapsed=07:23:26 avg_step=1.33s recent_step=1.42s eta=29:28:13 +2026-08-25 11:19:10 | INFO | train rank=0 | step=20060/100000 progress=20.1% loss=0.089050 lr=9.392638e-06 elapsed=07:23:40 avg_step=1.33s recent_step=1.41s eta=29:28:03 +2026-08-25 11:19:26 | INFO | train rank=0 | step=20070/100000 progress=20.1% loss=0.067963 lr=9.391848e-06 elapsed=07:23:56 avg_step=1.33s recent_step=1.59s eta=29:28:00 +2026-08-25 11:19:37 | INFO | train rank=0 | step=20080/100000 progress=20.1% loss=0.059364 lr=9.391058e-06 elapsed=07:24:07 avg_step=1.33s recent_step=1.14s eta=29:27:40 +2026-08-25 11:19:49 | INFO | train rank=0 | step=20090/100000 progress=20.1% loss=0.068634 lr=9.390267e-06 elapsed=07:24:19 avg_step=1.33s recent_step=1.21s eta=29:27:21 +2026-08-25 11:20:06 | INFO | train rank=0 | step=20100/100000 progress=20.1% loss=0.069977 lr=9.389475e-06 elapsed=07:24:36 avg_step=1.33s recent_step=1.64s eta=29:27:21 +2026-08-25 11:20:21 | INFO | train rank=0 | step=20110/100000 progress=20.1% loss=0.067047 lr=9.388683e-06 elapsed=07:24:50 avg_step=1.33s recent_step=1.48s eta=29:27:13 +2026-08-25 11:20:34 | INFO | train rank=0 | step=20120/100000 progress=20.1% loss=0.070465 lr=9.387891e-06 elapsed=07:25:04 avg_step=1.33s recent_step=1.31s eta=29:27:00 +2026-08-25 11:20:48 | INFO | train rank=0 | step=20130/100000 progress=20.1% loss=0.148010 lr=9.387098e-06 elapsed=07:25:18 avg_step=1.33s recent_step=1.42s eta=29:26:50 +2026-08-25 11:21:04 | INFO | train rank=0 | step=20140/100000 progress=20.1% loss=0.055878 lr=9.386304e-06 elapsed=07:25:34 avg_step=1.33s recent_step=1.66s eta=29:26:50 +2026-08-25 11:21:19 | INFO | train rank=0 | step=20150/100000 progress=20.2% loss=0.049438 lr=9.385510e-06 elapsed=07:25:49 avg_step=1.33s recent_step=1.47s eta=29:26:42 +2026-08-25 11:21:31 | INFO | train rank=0 | step=20160/100000 progress=20.2% loss=0.052505 lr=9.384716e-06 elapsed=07:26:01 avg_step=1.33s recent_step=1.23s eta=29:26:25 +2026-08-25 11:21:43 | INFO | train rank=0 | step=20170/100000 progress=20.2% loss=0.066467 lr=9.383921e-06 elapsed=07:26:13 avg_step=1.33s recent_step=1.17s eta=29:26:05 +2026-08-25 11:21:57 | INFO | train rank=0 | step=20180/100000 progress=20.2% loss=0.111450 lr=9.383126e-06 elapsed=07:26:27 avg_step=1.33s recent_step=1.37s eta=29:25:54 +2026-08-25 11:22:13 | INFO | train rank=0 | step=20190/100000 progress=20.2% loss=0.052902 lr=9.382330e-06 elapsed=07:26:43 avg_step=1.33s recent_step=1.61s eta=29:25:52 +2026-08-25 11:22:25 | INFO | train rank=0 | step=20200/100000 progress=20.2% loss=0.054474 lr=9.381533e-06 elapsed=07:26:55 avg_step=1.33s recent_step=1.20s eta=29:25:33 +2026-08-25 11:22:39 | INFO | train rank=0 | step=20210/100000 progress=20.2% loss=0.057007 lr=9.380737e-06 elapsed=07:27:09 avg_step=1.33s recent_step=1.38s eta=29:25:22 +2026-08-25 11:22:51 | INFO | train rank=0 | step=20220/100000 progress=20.2% loss=0.062866 lr=9.379939e-06 elapsed=07:27:21 avg_step=1.33s recent_step=1.20s eta=29:25:04 +2026-08-25 11:23:02 | INFO | train rank=0 | step=20230/100000 progress=20.2% loss=0.063293 lr=9.379142e-06 elapsed=07:27:32 avg_step=1.33s recent_step=1.11s eta=29:24:42 +2026-08-25 11:23:15 | INFO | train rank=0 | step=20240/100000 progress=20.2% loss=0.058792 lr=9.378343e-06 elapsed=07:27:45 avg_step=1.33s recent_step=1.31s eta=29:24:28 +2026-08-25 11:23:27 | INFO | train rank=0 | step=20250/100000 progress=20.2% loss=0.056824 lr=9.377545e-06 elapsed=07:27:57 avg_step=1.33s recent_step=1.25s eta=29:24:11 +2026-08-25 11:23:43 | INFO | train rank=0 | step=20260/100000 progress=20.3% loss=0.084442 lr=9.376745e-06 elapsed=07:28:13 avg_step=1.33s recent_step=1.53s eta=29:24:06 +2026-08-25 11:23:55 | INFO | train rank=0 | step=20270/100000 progress=20.3% loss=0.058960 lr=9.375946e-06 elapsed=07:28:25 avg_step=1.33s recent_step=1.24s eta=29:23:49 +2026-08-25 11:24:07 | INFO | train rank=0 | step=20280/100000 progress=20.3% loss=0.044640 lr=9.375146e-06 elapsed=07:28:37 avg_step=1.33s recent_step=1.21s eta=29:23:31 +2026-08-25 11:24:18 | INFO | train rank=0 | step=20290/100000 progress=20.3% loss=0.057114 lr=9.374345e-06 elapsed=07:28:48 avg_step=1.33s recent_step=1.14s eta=29:23:11 +2026-08-25 11:24:35 | INFO | train rank=0 | step=20300/100000 progress=20.3% loss=0.067169 lr=9.373544e-06 elapsed=07:29:04 avg_step=1.33s recent_step=1.61s eta=29:23:08 +2026-08-25 11:24:50 | INFO | train rank=0 | step=20310/100000 progress=20.3% loss=0.054138 lr=9.372742e-06 elapsed=07:29:20 avg_step=1.33s recent_step=1.55s eta=29:23:04 +2026-08-25 11:25:04 | INFO | train rank=0 | step=20320/100000 progress=20.3% loss=0.051941 lr=9.371940e-06 elapsed=07:29:33 avg_step=1.33s recent_step=1.35s eta=29:22:51 +2026-08-25 11:25:15 | INFO | train rank=0 | step=20330/100000 progress=20.3% loss=0.070129 lr=9.371138e-06 elapsed=07:29:45 avg_step=1.33s recent_step=1.18s eta=29:22:32 +2026-08-25 11:25:34 | INFO | train rank=0 | step=20340/100000 progress=20.3% loss=0.065002 lr=9.370335e-06 elapsed=07:30:04 avg_step=1.33s recent_step=1.86s eta=29:22:40 +2026-08-25 11:25:46 | INFO | train rank=0 | step=20350/100000 progress=20.3% loss=0.045395 lr=9.369531e-06 elapsed=07:30:16 avg_step=1.33s recent_step=1.21s eta=29:22:22 +2026-08-25 11:25:58 | INFO | train rank=0 | step=20360/100000 progress=20.4% loss=0.058121 lr=9.368727e-06 elapsed=07:30:28 avg_step=1.33s recent_step=1.21s eta=29:22:04 +2026-08-25 11:26:10 | INFO | train rank=0 | step=20370/100000 progress=20.4% loss=0.064011 lr=9.367923e-06 elapsed=07:30:39 avg_step=1.33s recent_step=1.14s eta=29:21:43 +2026-08-25 11:26:21 | INFO | train rank=0 | step=20380/100000 progress=20.4% loss=0.059761 lr=9.367118e-06 elapsed=07:30:51 avg_step=1.33s recent_step=1.17s eta=29:21:24 +2026-08-25 11:26:36 | INFO | train rank=0 | step=20390/100000 progress=20.4% loss=0.057953 lr=9.366312e-06 elapsed=07:31:06 avg_step=1.33s recent_step=1.45s eta=29:21:15 +2026-08-25 11:26:48 | INFO | train rank=0 | step=20400/100000 progress=20.4% loss=0.109619 lr=9.365506e-06 elapsed=07:31:18 avg_step=1.33s recent_step=1.22s eta=29:20:58 +2026-08-25 11:27:02 | INFO | train rank=0 | step=20410/100000 progress=20.4% loss=0.082703 lr=9.364700e-06 elapsed=07:31:32 avg_step=1.33s recent_step=1.39s eta=29:20:47 +2026-08-25 11:27:19 | INFO | train rank=0 | step=20420/100000 progress=20.4% loss=0.048187 lr=9.363893e-06 elapsed=07:31:49 avg_step=1.33s recent_step=1.68s eta=29:20:48 +2026-08-25 11:27:31 | INFO | train rank=0 | step=20430/100000 progress=20.4% loss=0.045288 lr=9.363086e-06 elapsed=07:32:00 avg_step=1.33s recent_step=1.18s eta=29:20:29 +2026-08-25 11:27:45 | INFO | train rank=0 | step=20440/100000 progress=20.4% loss=0.040726 lr=9.362278e-06 elapsed=07:32:15 avg_step=1.33s recent_step=1.43s eta=29:20:20 +2026-08-25 11:27:59 | INFO | train rank=0 | step=20450/100000 progress=20.4% loss=0.129822 lr=9.361470e-06 elapsed=07:32:29 avg_step=1.33s recent_step=1.45s eta=29:20:11 +2026-08-25 11:28:14 | INFO | train rank=0 | step=20460/100000 progress=20.5% loss=0.065216 lr=9.360661e-06 elapsed=07:32:44 avg_step=1.33s recent_step=1.46s eta=29:20:03 +2026-08-25 11:28:28 | INFO | train rank=0 | step=20470/100000 progress=20.5% loss=0.047272 lr=9.359852e-06 elapsed=07:32:58 avg_step=1.33s recent_step=1.43s eta=29:19:54 +2026-08-25 11:28:43 | INFO | train rank=0 | step=20480/100000 progress=20.5% loss=0.089539 lr=9.359042e-06 elapsed=07:33:13 avg_step=1.33s recent_step=1.47s eta=29:19:46 +2026-08-25 11:28:54 | INFO | train rank=0 | step=20490/100000 progress=20.5% loss=0.110809 lr=9.358232e-06 elapsed=07:33:24 avg_step=1.33s recent_step=1.15s eta=29:19:25 +2026-08-25 11:29:06 | INFO | train rank=0 | step=20500/100000 progress=20.5% loss=0.052994 lr=9.357421e-06 elapsed=07:33:35 avg_step=1.33s recent_step=1.12s eta=29:19:04 +2026-08-25 11:29:17 | INFO | train rank=0 | step=20510/100000 progress=20.5% loss=0.103363 lr=9.356610e-06 elapsed=07:33:47 avg_step=1.33s recent_step=1.18s eta=29:18:45 +2026-08-25 11:29:29 | INFO | train rank=0 | step=20520/100000 progress=20.5% loss=0.036530 lr=9.355798e-06 elapsed=07:33:59 avg_step=1.33s recent_step=1.15s eta=29:18:25 +2026-08-25 11:29:44 | INFO | train rank=0 | step=20530/100000 progress=20.5% loss=0.059860 lr=9.354986e-06 elapsed=07:34:14 avg_step=1.33s recent_step=1.54s eta=29:18:20 +2026-08-25 11:30:00 | INFO | train rank=0 | step=20540/100000 progress=20.5% loss=0.077209 lr=9.354174e-06 elapsed=07:34:30 avg_step=1.33s recent_step=1.58s eta=29:18:16 +2026-08-25 11:30:11 | INFO | train rank=0 | step=20550/100000 progress=20.5% loss=0.048935 lr=9.353361e-06 elapsed=07:34:41 avg_step=1.33s recent_step=1.13s eta=29:17:56 +2026-08-25 11:30:24 | INFO | train rank=0 | step=20560/100000 progress=20.6% loss=0.058685 lr=9.352547e-06 elapsed=07:34:54 avg_step=1.33s recent_step=1.31s eta=29:17:42 +2026-08-25 11:30:39 | INFO | train rank=0 | step=20570/100000 progress=20.6% loss=0.070435 lr=9.351733e-06 elapsed=07:35:09 avg_step=1.33s recent_step=1.48s eta=29:17:34 +2026-08-25 11:30:54 | INFO | train rank=0 | step=20580/100000 progress=20.6% loss=0.071899 lr=9.350919e-06 elapsed=07:35:24 avg_step=1.33s recent_step=1.51s eta=29:17:28 +2026-08-25 11:31:06 | INFO | train rank=0 | step=20590/100000 progress=20.6% loss=0.059204 lr=9.350104e-06 elapsed=07:35:36 avg_step=1.33s recent_step=1.16s eta=29:17:08 +2026-08-25 11:31:21 | INFO | train rank=0 | step=20600/100000 progress=20.6% loss=0.081558 lr=9.349288e-06 elapsed=07:35:51 avg_step=1.33s recent_step=1.49s eta=29:17:01 +2026-08-25 11:31:34 | INFO | train rank=0 | step=20610/100000 progress=20.6% loss=0.093018 lr=9.348473e-06 elapsed=07:36:04 avg_step=1.33s recent_step=1.32s eta=29:16:48 +2026-08-25 11:31:52 | INFO | train rank=0 | step=20620/100000 progress=20.6% loss=0.059967 lr=9.347656e-06 elapsed=07:36:21 avg_step=1.33s recent_step=1.76s eta=29:16:51 +2026-08-25 11:32:07 | INFO | train rank=0 | step=20630/100000 progress=20.6% loss=0.046982 lr=9.346839e-06 elapsed=07:36:37 avg_step=1.33s recent_step=1.55s eta=29:16:46 +2026-08-25 11:32:19 | INFO | train rank=0 | step=20640/100000 progress=20.6% loss=0.084656 lr=9.346022e-06 elapsed=07:36:48 avg_step=1.33s recent_step=1.15s eta=29:16:26 +2026-08-25 11:32:30 | INFO | train rank=0 | step=20650/100000 progress=20.6% loss=0.140320 lr=9.345204e-06 elapsed=07:37:00 avg_step=1.33s recent_step=1.15s eta=29:16:06 +2026-08-25 11:32:45 | INFO | train rank=0 | step=20660/100000 progress=20.7% loss=0.073486 lr=9.344386e-06 elapsed=07:37:15 avg_step=1.33s recent_step=1.49s eta=29:15:59 +2026-08-25 11:32:57 | INFO | train rank=0 | step=20670/100000 progress=20.7% loss=0.076721 lr=9.343567e-06 elapsed=07:37:27 avg_step=1.33s recent_step=1.20s eta=29:15:40 +2026-08-25 11:33:08 | INFO | train rank=0 | step=20680/100000 progress=20.7% loss=0.060394 lr=9.342748e-06 elapsed=07:37:38 avg_step=1.33s recent_step=1.13s eta=29:15:19 +2026-08-25 11:33:22 | INFO | train rank=0 | step=20690/100000 progress=20.7% loss=0.058426 lr=9.341928e-06 elapsed=07:37:52 avg_step=1.33s recent_step=1.41s eta=29:15:09 +2026-08-25 11:33:40 | INFO | train rank=0 | step=20700/100000 progress=20.7% loss=0.061768 lr=9.341108e-06 elapsed=07:38:10 avg_step=1.33s recent_step=1.81s eta=29:15:14 +2026-08-25 11:33:51 | INFO | train rank=0 | step=20710/100000 progress=20.7% loss=0.066010 lr=9.340287e-06 elapsed=07:38:21 avg_step=1.33s recent_step=1.11s eta=29:14:53 +2026-08-25 11:34:04 | INFO | train rank=0 | step=20720/100000 progress=20.7% loss=0.059601 lr=9.339466e-06 elapsed=07:38:34 avg_step=1.33s recent_step=1.29s eta=29:14:38 +2026-08-25 11:34:20 | INFO | train rank=0 | step=20730/100000 progress=20.7% loss=0.077515 lr=9.338645e-06 elapsed=07:38:49 avg_step=1.33s recent_step=1.51s eta=29:14:32 +2026-08-25 11:34:31 | INFO | train rank=0 | step=20740/100000 progress=20.7% loss=0.100891 lr=9.337823e-06 elapsed=07:39:01 avg_step=1.33s recent_step=1.18s eta=29:14:13 +2026-08-25 11:34:46 | INFO | train rank=0 | step=20750/100000 progress=20.8% loss=0.083679 lr=9.337000e-06 elapsed=07:39:16 avg_step=1.33s recent_step=1.43s eta=29:14:04 +2026-08-25 11:34:57 | INFO | train rank=0 | step=20760/100000 progress=20.8% loss=0.063995 lr=9.336177e-06 elapsed=07:39:27 avg_step=1.33s recent_step=1.15s eta=29:13:44 +2026-08-25 11:35:09 | INFO | train rank=0 | step=20770/100000 progress=20.8% loss=0.038513 lr=9.335354e-06 elapsed=07:39:39 avg_step=1.33s recent_step=1.17s eta=29:13:24 +2026-08-25 11:35:25 | INFO | train rank=0 | step=20780/100000 progress=20.8% loss=0.047028 lr=9.334530e-06 elapsed=07:39:55 avg_step=1.33s recent_step=1.67s eta=29:13:24 +2026-08-25 11:35:42 | INFO | train rank=0 | step=20790/100000 progress=20.8% loss=0.058548 lr=9.333705e-06 elapsed=07:40:12 avg_step=1.33s recent_step=1.64s eta=29:13:22 +2026-08-25 11:35:54 | INFO | train rank=0 | step=20800/100000 progress=20.8% loss=0.052597 lr=9.332880e-06 elapsed=07:40:24 avg_step=1.33s recent_step=1.18s eta=29:13:03 +2026-08-25 11:36:06 | INFO | train rank=0 | step=20810/100000 progress=20.8% loss=0.089966 lr=9.332055e-06 elapsed=07:40:36 avg_step=1.33s recent_step=1.21s eta=29:12:46 +2026-08-25 11:36:20 | INFO | train rank=0 | step=20820/100000 progress=20.8% loss=0.046272 lr=9.331229e-06 elapsed=07:40:50 avg_step=1.33s recent_step=1.41s eta=29:12:35 +2026-08-25 11:36:32 | INFO | train rank=0 | step=20830/100000 progress=20.8% loss=0.067139 lr=9.330403e-06 elapsed=07:41:02 avg_step=1.33s recent_step=1.18s eta=29:12:17 +2026-08-25 11:36:44 | INFO | train rank=0 | step=20840/100000 progress=20.8% loss=0.040527 lr=9.329576e-06 elapsed=07:41:13 avg_step=1.33s recent_step=1.18s eta=29:11:58 +2026-08-25 11:36:59 | INFO | train rank=0 | step=20850/100000 progress=20.8% loss=0.063660 lr=9.328748e-06 elapsed=07:41:28 avg_step=1.33s recent_step=1.50s eta=29:11:51 +2026-08-25 11:37:10 | INFO | train rank=0 | step=20860/100000 progress=20.9% loss=0.023506 lr=9.327921e-06 elapsed=07:41:40 avg_step=1.33s recent_step=1.13s eta=29:11:31 +2026-08-25 11:37:23 | INFO | train rank=0 | step=20870/100000 progress=20.9% loss=0.068207 lr=9.327092e-06 elapsed=07:41:53 avg_step=1.33s recent_step=1.31s eta=29:11:16 +2026-08-25 11:37:39 | INFO | train rank=0 | step=20880/100000 progress=20.9% loss=0.068970 lr=9.326264e-06 elapsed=07:42:09 avg_step=1.33s recent_step=1.63s eta=29:11:14 +2026-08-25 11:37:52 | INFO | train rank=0 | step=20890/100000 progress=20.9% loss=0.058823 lr=9.325435e-06 elapsed=07:42:22 avg_step=1.33s recent_step=1.25s eta=29:10:58 +2026-08-25 11:38:07 | INFO | train rank=0 | step=20900/100000 progress=20.9% loss=0.065552 lr=9.324605e-06 elapsed=07:42:36 avg_step=1.33s recent_step=1.49s eta=29:10:51 +2026-08-25 11:38:18 | INFO | train rank=0 | step=20910/100000 progress=20.9% loss=0.073090 lr=9.323775e-06 elapsed=07:42:48 avg_step=1.33s recent_step=1.16s eta=29:10:31 +2026-08-25 11:38:29 | INFO | train rank=0 | step=20920/100000 progress=20.9% loss=0.055222 lr=9.322944e-06 elapsed=07:42:59 avg_step=1.33s recent_step=1.11s eta=29:10:10 +2026-08-25 11:38:41 | INFO | train rank=0 | step=20930/100000 progress=20.9% loss=0.055725 lr=9.322113e-06 elapsed=07:43:11 avg_step=1.33s recent_step=1.20s eta=29:09:52 +2026-08-25 11:38:55 | INFO | train rank=0 | step=20940/100000 progress=20.9% loss=0.063690 lr=9.321282e-06 elapsed=07:43:25 avg_step=1.33s recent_step=1.42s eta=29:09:42 +2026-08-25 11:39:07 | INFO | train rank=0 | step=20950/100000 progress=20.9% loss=0.062317 lr=9.320450e-06 elapsed=07:43:37 avg_step=1.33s recent_step=1.17s eta=29:09:23 +2026-08-25 11:39:21 | INFO | train rank=0 | step=20960/100000 progress=21.0% loss=0.107971 lr=9.319617e-06 elapsed=07:43:51 avg_step=1.33s recent_step=1.36s eta=29:09:11 +2026-08-25 11:39:32 | INFO | train rank=0 | step=20970/100000 progress=21.0% loss=0.079041 lr=9.318784e-06 elapsed=07:44:02 avg_step=1.33s recent_step=1.14s eta=29:08:51 +2026-08-25 11:39:44 | INFO | train rank=0 | step=20980/100000 progress=21.0% loss=0.051453 lr=9.317951e-06 elapsed=07:44:14 avg_step=1.33s recent_step=1.18s eta=29:08:32 +2026-08-25 11:39:58 | INFO | train rank=0 | step=20990/100000 progress=21.0% loss=0.097168 lr=9.317117e-06 elapsed=07:44:28 avg_step=1.33s recent_step=1.38s eta=29:08:20 +2026-08-25 11:40:10 | INFO | train rank=0 | step=21000/100000 progress=21.0% loss=0.049530 lr=9.316282e-06 elapsed=07:44:40 avg_step=1.33s recent_step=1.18s eta=29:08:02 +2026-08-25 11:40:21 | INFO | train rank=0 | step=21010/100000 progress=21.0% loss=0.064972 lr=9.315448e-06 elapsed=07:44:51 avg_step=1.33s recent_step=1.18s eta=29:07:43 +2026-08-25 11:40:33 | INFO | train rank=0 | step=21020/100000 progress=21.0% loss=0.123962 lr=9.314612e-06 elapsed=07:45:03 avg_step=1.33s recent_step=1.12s eta=29:07:22 +2026-08-25 11:40:45 | INFO | train rank=0 | step=21030/100000 progress=21.0% loss=0.049774 lr=9.313776e-06 elapsed=07:45:15 avg_step=1.33s recent_step=1.21s eta=29:07:04 +2026-08-25 11:40:57 | INFO | train rank=0 | step=21040/100000 progress=21.0% loss=0.053192 lr=9.312940e-06 elapsed=07:45:27 avg_step=1.33s recent_step=1.27s eta=29:06:49 +2026-08-25 11:41:11 | INFO | train rank=0 | step=21050/100000 progress=21.1% loss=0.083313 lr=9.312103e-06 elapsed=07:45:41 avg_step=1.33s recent_step=1.37s eta=29:06:37 +2026-08-25 11:41:26 | INFO | train rank=0 | step=21060/100000 progress=21.1% loss=0.062042 lr=9.311266e-06 elapsed=07:45:56 avg_step=1.33s recent_step=1.49s eta=29:06:30 +2026-08-25 11:41:42 | INFO | train rank=0 | step=21070/100000 progress=21.1% loss=0.076416 lr=9.310429e-06 elapsed=07:46:12 avg_step=1.33s recent_step=1.56s eta=29:06:25 +2026-08-25 11:41:56 | INFO | train rank=0 | step=21080/100000 progress=21.1% loss=0.056610 lr=9.309590e-06 elapsed=07:46:26 avg_step=1.33s recent_step=1.45s eta=29:06:16 +2026-08-25 11:42:08 | INFO | train rank=0 | step=21090/100000 progress=21.1% loss=0.071625 lr=9.308752e-06 elapsed=07:46:38 avg_step=1.33s recent_step=1.15s eta=29:05:57 +2026-08-25 11:42:22 | INFO | train rank=0 | step=21100/100000 progress=21.1% loss=0.067474 lr=9.307913e-06 elapsed=07:46:52 avg_step=1.33s recent_step=1.47s eta=29:05:49 +2026-08-25 11:42:34 | INFO | train rank=0 | step=21110/100000 progress=21.1% loss=0.056778 lr=9.307073e-06 elapsed=07:47:04 avg_step=1.33s recent_step=1.13s eta=29:05:28 +2026-08-25 11:42:48 | INFO | train rank=0 | step=21120/100000 progress=21.1% loss=0.033432 lr=9.306233e-06 elapsed=07:47:18 avg_step=1.33s recent_step=1.40s eta=29:05:18 +2026-08-25 11:43:04 | INFO | train rank=0 | step=21130/100000 progress=21.1% loss=0.048401 lr=9.305393e-06 elapsed=07:47:34 avg_step=1.33s recent_step=1.67s eta=29:05:17 +2026-08-25 11:43:16 | INFO | train rank=0 | step=21140/100000 progress=21.1% loss=0.051895 lr=9.304552e-06 elapsed=07:47:46 avg_step=1.33s recent_step=1.16s eta=29:04:58 +2026-08-25 11:43:29 | INFO | train rank=0 | step=21150/100000 progress=21.1% loss=0.045334 lr=9.303710e-06 elapsed=07:47:59 avg_step=1.33s recent_step=1.30s eta=29:04:43 +2026-08-25 11:43:42 | INFO | train rank=0 | step=21160/100000 progress=21.2% loss=0.073486 lr=9.302868e-06 elapsed=07:48:11 avg_step=1.33s recent_step=1.25s eta=29:04:27 +2026-08-25 11:43:57 | INFO | train rank=0 | step=21170/100000 progress=21.2% loss=0.065369 lr=9.302026e-06 elapsed=07:48:27 avg_step=1.33s recent_step=1.54s eta=29:04:22 +2026-08-25 11:44:14 | INFO | train rank=0 | step=21180/100000 progress=21.2% loss=0.030884 lr=9.301183e-06 elapsed=07:48:43 avg_step=1.33s recent_step=1.67s eta=29:04:21 +2026-08-25 11:44:25 | INFO | train rank=0 | step=21190/100000 progress=21.2% loss=0.044174 lr=9.300340e-06 elapsed=07:48:55 avg_step=1.33s recent_step=1.16s eta=29:04:02 +2026-08-25 11:44:37 | INFO | train rank=0 | step=21200/100000 progress=21.2% loss=0.063873 lr=9.299496e-06 elapsed=07:49:07 avg_step=1.33s recent_step=1.22s eta=29:03:44 +2026-08-25 11:44:51 | INFO | train rank=0 | step=21210/100000 progress=21.2% loss=0.070419 lr=9.298652e-06 elapsed=07:49:21 avg_step=1.33s recent_step=1.39s eta=29:03:33 +2026-08-25 11:45:03 | INFO | train rank=0 | step=21220/100000 progress=21.2% loss=0.044876 lr=9.297807e-06 elapsed=07:49:33 avg_step=1.33s recent_step=1.18s eta=29:03:15 +2026-08-25 11:45:15 | INFO | train rank=0 | step=21230/100000 progress=21.2% loss=0.052948 lr=9.296962e-06 elapsed=07:49:45 avg_step=1.33s recent_step=1.23s eta=29:02:58 +2026-08-25 11:45:31 | INFO | train rank=0 | step=21240/100000 progress=21.2% loss=0.087921 lr=9.296116e-06 elapsed=07:50:01 avg_step=1.33s recent_step=1.53s eta=29:02:52 +2026-08-25 11:45:45 | INFO | train rank=0 | step=21250/100000 progress=21.2% loss=0.050095 lr=9.295270e-06 elapsed=07:50:15 avg_step=1.33s recent_step=1.42s eta=29:02:42 +2026-08-25 11:46:00 | INFO | train rank=0 | step=21260/100000 progress=21.3% loss=0.104309 lr=9.294423e-06 elapsed=07:50:30 avg_step=1.33s recent_step=1.54s eta=29:02:37 +2026-08-25 11:46:14 | INFO | train rank=0 | step=21270/100000 progress=21.3% loss=0.073944 lr=9.293576e-06 elapsed=07:50:44 avg_step=1.33s recent_step=1.34s eta=29:02:24 +2026-08-25 11:46:25 | INFO | train rank=0 | step=21280/100000 progress=21.3% loss=0.063019 lr=9.292728e-06 elapsed=07:50:55 avg_step=1.33s recent_step=1.15s eta=29:02:04 +2026-08-25 11:46:38 | INFO | train rank=0 | step=21290/100000 progress=21.3% loss=0.069275 lr=9.291880e-06 elapsed=07:51:08 avg_step=1.33s recent_step=1.31s eta=29:01:50 +2026-08-25 11:46:51 | INFO | train rank=0 | step=21300/100000 progress=21.3% loss=0.039253 lr=9.291032e-06 elapsed=07:51:21 avg_step=1.33s recent_step=1.32s eta=29:01:36 +2026-08-25 11:47:09 | INFO | train rank=0 | step=21310/100000 progress=21.3% loss=0.054901 lr=9.290183e-06 elapsed=07:51:39 avg_step=1.33s recent_step=1.74s eta=29:01:38 +2026-08-25 11:47:20 | INFO | train rank=0 | step=21320/100000 progress=21.3% loss=0.082794 lr=9.289334e-06 elapsed=07:51:50 avg_step=1.33s recent_step=1.15s eta=29:01:18 +2026-08-25 11:47:36 | INFO | train rank=0 | step=21330/100000 progress=21.3% loss=0.075775 lr=9.288484e-06 elapsed=07:52:06 avg_step=1.33s recent_step=1.60s eta=29:01:15 +2026-08-25 11:47:48 | INFO | train rank=0 | step=21340/100000 progress=21.3% loss=0.034409 lr=9.287633e-06 elapsed=07:52:17 avg_step=1.33s recent_step=1.12s eta=29:00:54 +2026-08-25 11:48:03 | INFO | train rank=0 | step=21350/100000 progress=21.3% loss=0.091614 lr=9.286782e-06 elapsed=07:52:33 avg_step=1.33s recent_step=1.51s eta=29:00:47 +2026-08-25 11:48:15 | INFO | train rank=0 | step=21360/100000 progress=21.4% loss=0.056396 lr=9.285931e-06 elapsed=07:52:44 avg_step=1.33s recent_step=1.19s eta=29:00:29 +2026-08-25 11:48:26 | INFO | train rank=0 | step=21370/100000 progress=21.4% loss=0.048737 lr=9.285079e-06 elapsed=07:52:56 avg_step=1.33s recent_step=1.13s eta=29:00:08 +2026-08-25 11:48:43 | INFO | train rank=0 | step=21380/100000 progress=21.4% loss=0.067657 lr=9.284227e-06 elapsed=07:53:13 avg_step=1.33s recent_step=1.74s eta=29:00:10 +2026-08-25 11:48:59 | INFO | train rank=0 | step=21390/100000 progress=21.4% loss=0.059875 lr=9.283374e-06 elapsed=07:53:29 avg_step=1.33s recent_step=1.57s eta=29:00:06 +2026-08-25 11:49:16 | INFO | train rank=0 | step=21400/100000 progress=21.4% loss=0.045120 lr=9.282521e-06 elapsed=07:53:45 avg_step=1.33s recent_step=1.67s eta=29:00:05 +2026-08-25 11:49:27 | INFO | train rank=0 | step=21410/100000 progress=21.4% loss=0.048019 lr=9.281668e-06 elapsed=07:53:57 avg_step=1.33s recent_step=1.19s eta=28:59:47 +2026-08-25 11:49:39 | INFO | train rank=0 | step=21420/100000 progress=21.4% loss=0.097168 lr=9.280813e-06 elapsed=07:54:09 avg_step=1.33s recent_step=1.12s eta=28:59:26 +2026-08-25 11:49:52 | INFO | train rank=0 | step=21430/100000 progress=21.4% loss=0.088013 lr=9.279959e-06 elapsed=07:54:22 avg_step=1.33s recent_step=1.34s eta=28:59:13 +2026-08-25 11:50:04 | INFO | train rank=0 | step=21440/100000 progress=21.4% loss=0.055771 lr=9.279104e-06 elapsed=07:54:34 avg_step=1.33s recent_step=1.20s eta=28:58:55 +2026-08-25 11:50:17 | INFO | train rank=0 | step=21450/100000 progress=21.4% loss=0.025864 lr=9.278248e-06 elapsed=07:54:47 avg_step=1.33s recent_step=1.27s eta=28:58:39 +2026-08-25 11:50:30 | INFO | train rank=0 | step=21460/100000 progress=21.5% loss=0.056931 lr=9.277392e-06 elapsed=07:55:00 avg_step=1.33s recent_step=1.29s eta=28:58:25 +2026-08-25 11:50:45 | INFO | train rank=0 | step=21470/100000 progress=21.5% loss=0.034935 lr=9.276536e-06 elapsed=07:55:15 avg_step=1.33s recent_step=1.52s eta=28:58:18 +2026-08-25 11:51:01 | INFO | train rank=0 | step=21480/100000 progress=21.5% loss=0.059479 lr=9.275679e-06 elapsed=07:55:31 avg_step=1.33s recent_step=1.65s eta=28:58:17 +2026-08-25 11:51:13 | INFO | train rank=0 | step=21490/100000 progress=21.5% loss=0.104614 lr=9.274821e-06 elapsed=07:55:43 avg_step=1.33s recent_step=1.21s eta=28:58:00 +2026-08-25 11:51:25 | INFO | train rank=0 | step=21500/100000 progress=21.5% loss=0.042542 lr=9.273964e-06 elapsed=07:55:55 avg_step=1.33s recent_step=1.13s eta=28:57:39 +2026-08-25 11:51:41 | INFO | train rank=0 | step=21510/100000 progress=21.5% loss=0.054535 lr=9.273105e-06 elapsed=07:56:11 avg_step=1.33s recent_step=1.61s eta=28:57:36 +2026-08-25 11:51:53 | INFO | train rank=0 | step=21520/100000 progress=21.5% loss=0.050903 lr=9.272246e-06 elapsed=07:56:23 avg_step=1.33s recent_step=1.18s eta=28:57:17 +2026-08-25 11:52:07 | INFO | train rank=0 | step=21530/100000 progress=21.5% loss=0.072540 lr=9.271387e-06 elapsed=07:56:37 avg_step=1.33s recent_step=1.41s eta=28:57:07 +2026-08-25 11:52:19 | INFO | train rank=0 | step=21540/100000 progress=21.5% loss=0.042236 lr=9.270527e-06 elapsed=07:56:49 avg_step=1.33s recent_step=1.25s eta=28:56:51 +2026-08-25 11:52:31 | INFO | train rank=0 | step=21550/100000 progress=21.6% loss=0.064255 lr=9.269667e-06 elapsed=07:57:01 avg_step=1.33s recent_step=1.17s eta=28:56:32 +2026-08-25 11:52:42 | INFO | train rank=0 | step=21560/100000 progress=21.6% loss=0.069061 lr=9.268807e-06 elapsed=07:57:12 avg_step=1.33s recent_step=1.10s eta=28:56:10 +2026-08-25 11:52:54 | INFO | train rank=0 | step=21570/100000 progress=21.6% loss=0.044739 lr=9.267945e-06 elapsed=07:57:24 avg_step=1.33s recent_step=1.22s eta=28:55:53 +2026-08-25 11:53:05 | INFO | train rank=0 | step=21580/100000 progress=21.6% loss=0.067383 lr=9.267084e-06 elapsed=07:57:35 avg_step=1.33s recent_step=1.10s eta=28:55:32 +2026-08-25 11:53:24 | INFO | train rank=0 | step=21590/100000 progress=21.6% loss=0.048615 lr=9.266222e-06 elapsed=07:57:53 avg_step=1.33s recent_step=1.84s eta=28:55:37 +2026-08-25 11:53:39 | INFO | train rank=0 | step=21600/100000 progress=21.6% loss=0.069916 lr=9.265359e-06 elapsed=07:58:08 avg_step=1.33s recent_step=1.50s eta=28:55:30 +2026-08-25 11:53:51 | INFO | train rank=0 | step=21610/100000 progress=21.6% loss=0.048767 lr=9.264496e-06 elapsed=07:58:21 avg_step=1.33s recent_step=1.27s eta=28:55:14 +2026-08-25 11:54:03 | INFO | train rank=0 | step=21620/100000 progress=21.6% loss=0.071106 lr=9.263633e-06 elapsed=07:58:33 avg_step=1.33s recent_step=1.21s eta=28:54:57 +2026-08-25 11:54:14 | INFO | train rank=0 | step=21630/100000 progress=21.6% loss=0.044464 lr=9.262769e-06 elapsed=07:58:44 avg_step=1.33s recent_step=1.10s eta=28:54:35 +2026-08-25 11:54:26 | INFO | train rank=0 | step=21640/100000 progress=21.6% loss=0.057007 lr=9.261904e-06 elapsed=07:58:56 avg_step=1.33s recent_step=1.15s eta=28:54:16 +2026-08-25 11:54:44 | INFO | train rank=0 | step=21650/100000 progress=21.6% loss=0.061157 lr=9.261040e-06 elapsed=07:59:14 avg_step=1.33s recent_step=1.79s eta=28:54:19 +2026-08-25 11:54:55 | INFO | train rank=0 | step=21660/100000 progress=21.7% loss=0.062698 lr=9.260174e-06 elapsed=07:59:25 avg_step=1.33s recent_step=1.11s eta=28:53:58 +2026-08-25 11:55:10 | INFO | train rank=0 | step=21670/100000 progress=21.7% loss=0.075439 lr=9.259308e-06 elapsed=07:59:39 avg_step=1.33s recent_step=1.46s eta=28:53:50 +2026-08-25 11:55:24 | INFO | train rank=0 | step=21680/100000 progress=21.7% loss=0.068054 lr=9.258442e-06 elapsed=07:59:54 avg_step=1.33s recent_step=1.43s eta=28:53:40 +2026-08-25 11:55:36 | INFO | train rank=0 | step=21690/100000 progress=21.7% loss=0.080536 lr=9.257575e-06 elapsed=08:00:05 avg_step=1.33s recent_step=1.16s eta=28:53:21 +2026-08-25 11:55:47 | INFO | train rank=0 | step=21700/100000 progress=21.7% loss=0.073425 lr=9.256708e-06 elapsed=08:00:17 avg_step=1.33s recent_step=1.18s eta=28:53:02 +2026-08-25 11:56:04 | INFO | train rank=0 | step=21710/100000 progress=21.7% loss=0.088806 lr=9.255841e-06 elapsed=08:00:33 avg_step=1.33s recent_step=1.63s eta=28:53:00 +2026-08-25 11:56:15 | INFO | train rank=0 | step=21720/100000 progress=21.7% loss=0.078362 lr=9.254972e-06 elapsed=08:00:45 avg_step=1.33s recent_step=1.15s eta=28:52:40 +2026-08-25 11:56:27 | INFO | train rank=0 | step=21730/100000 progress=21.7% loss=0.059113 lr=9.254104e-06 elapsed=08:00:57 avg_step=1.33s recent_step=1.18s eta=28:52:22 +2026-08-25 11:56:38 | INFO | train rank=0 | step=21740/100000 progress=21.7% loss=0.057892 lr=9.253235e-06 elapsed=08:01:08 avg_step=1.33s recent_step=1.12s eta=28:52:01 +2026-08-25 11:56:55 | INFO | train rank=0 | step=21750/100000 progress=21.8% loss=0.097046 lr=9.252365e-06 elapsed=08:01:25 avg_step=1.33s recent_step=1.72s eta=28:52:02 +2026-08-25 11:57:08 | INFO | train rank=0 | step=21760/100000 progress=21.8% loss=0.052650 lr=9.251495e-06 elapsed=08:01:37 avg_step=1.33s recent_step=1.23s eta=28:51:45 +2026-08-25 11:57:22 | INFO | train rank=0 | step=21770/100000 progress=21.8% loss=0.050293 lr=9.250625e-06 elapsed=08:01:52 avg_step=1.33s recent_step=1.49s eta=28:51:37 +2026-08-25 11:57:34 | INFO | train rank=0 | step=21780/100000 progress=21.8% loss=0.073486 lr=9.249754e-06 elapsed=08:02:04 avg_step=1.33s recent_step=1.21s eta=28:51:19 +2026-08-25 11:57:49 | INFO | train rank=0 | step=21790/100000 progress=21.8% loss=0.050568 lr=9.248882e-06 elapsed=08:02:18 avg_step=1.33s recent_step=1.41s eta=28:51:09 +2026-08-25 11:58:00 | INFO | train rank=0 | step=21800/100000 progress=21.8% loss=0.090454 lr=9.248011e-06 elapsed=08:02:30 avg_step=1.33s recent_step=1.18s eta=28:50:51 +2026-08-25 11:58:12 | INFO | train rank=0 | step=21810/100000 progress=21.8% loss=0.073975 lr=9.247138e-06 elapsed=08:02:42 avg_step=1.33s recent_step=1.13s eta=28:50:30 +2026-08-25 11:58:23 | INFO | train rank=0 | step=21820/100000 progress=21.8% loss=0.059326 lr=9.246266e-06 elapsed=08:02:53 avg_step=1.33s recent_step=1.12s eta=28:50:09 +2026-08-25 11:58:34 | INFO | train rank=0 | step=21830/100000 progress=21.8% loss=0.074677 lr=9.245392e-06 elapsed=08:03:04 avg_step=1.33s recent_step=1.15s eta=28:49:50 +2026-08-25 11:58:53 | INFO | train rank=0 | step=21840/100000 progress=21.8% loss=0.047653 lr=9.244519e-06 elapsed=08:03:23 avg_step=1.33s recent_step=1.84s eta=28:49:55 +2026-08-25 11:59:04 | INFO | train rank=0 | step=21850/100000 progress=21.9% loss=0.066162 lr=9.243644e-06 elapsed=08:03:34 avg_step=1.33s recent_step=1.14s eta=28:49:35 +2026-08-25 11:59:20 | INFO | train rank=0 | step=21860/100000 progress=21.9% loss=0.077118 lr=9.242770e-06 elapsed=08:03:50 avg_step=1.33s recent_step=1.55s eta=28:49:29 +2026-08-25 11:59:31 | INFO | train rank=0 | step=21870/100000 progress=21.9% loss=0.056534 lr=9.241895e-06 elapsed=08:04:01 avg_step=1.33s recent_step=1.12s eta=28:49:09 +2026-08-25 11:59:43 | INFO | train rank=0 | step=21880/100000 progress=21.9% loss=0.091949 lr=9.241019e-06 elapsed=08:04:12 avg_step=1.33s recent_step=1.17s eta=28:48:50 +2026-08-25 11:59:57 | INFO | train rank=0 | step=21890/100000 progress=21.9% loss=0.067932 lr=9.240143e-06 elapsed=08:04:27 avg_step=1.33s recent_step=1.42s eta=28:48:40 +2026-08-25 12:00:13 | INFO | train rank=0 | step=21900/100000 progress=21.9% loss=0.059296 lr=9.239267e-06 elapsed=08:04:42 avg_step=1.33s recent_step=1.58s eta=28:48:35 +2026-08-25 12:00:30 | INFO | train rank=0 | step=21910/100000 progress=21.9% loss=0.046448 lr=9.238390e-06 elapsed=08:04:59 avg_step=1.33s recent_step=1.70s eta=28:48:35 +2026-08-25 12:00:41 | INFO | train rank=0 | step=21920/100000 progress=21.9% loss=0.055786 lr=9.237512e-06 elapsed=08:05:11 avg_step=1.33s recent_step=1.11s eta=28:48:14 +2026-08-25 12:00:53 | INFO | train rank=0 | step=21930/100000 progress=21.9% loss=0.047180 lr=9.236634e-06 elapsed=08:05:23 avg_step=1.33s recent_step=1.23s eta=28:47:57 +2026-08-25 12:01:08 | INFO | train rank=0 | step=21940/100000 progress=21.9% loss=0.062225 lr=9.235756e-06 elapsed=08:05:38 avg_step=1.33s recent_step=1.53s eta=28:47:51 +2026-08-25 12:01:19 | INFO | train rank=0 | step=21950/100000 progress=21.9% loss=0.060272 lr=9.234877e-06 elapsed=08:05:49 avg_step=1.33s recent_step=1.10s eta=28:47:30 +2026-08-25 12:01:31 | INFO | train rank=0 | step=21960/100000 progress=22.0% loss=0.068237 lr=9.233998e-06 elapsed=08:06:01 avg_step=1.33s recent_step=1.17s eta=28:47:11 +2026-08-25 12:01:42 | INFO | train rank=0 | step=21970/100000 progress=22.0% loss=0.040985 lr=9.233118e-06 elapsed=08:06:12 avg_step=1.33s recent_step=1.13s eta=28:46:51 +2026-08-25 12:01:54 | INFO | train rank=0 | step=21980/100000 progress=22.0% loss=0.062119 lr=9.232238e-06 elapsed=08:06:24 avg_step=1.33s recent_step=1.14s eta=28:46:31 +2026-08-25 12:02:10 | INFO | train rank=0 | step=21990/100000 progress=22.0% loss=0.059357 lr=9.231357e-06 elapsed=08:06:40 avg_step=1.33s recent_step=1.67s eta=28:46:30 +2026-08-25 12:02:22 | INFO | train rank=0 | step=22000/100000 progress=22.0% loss=0.057678 lr=9.230476e-06 elapsed=08:06:52 avg_step=1.33s recent_step=1.19s eta=28:46:12 +2026-08-25 12:02:35 | INFO | train rank=0 | step=22010/100000 progress=22.0% loss=0.043289 lr=9.229595e-06 elapsed=08:07:05 avg_step=1.33s recent_step=1.24s eta=28:45:55 +2026-08-25 12:02:47 | INFO | train rank=0 | step=22020/100000 progress=22.0% loss=0.067352 lr=9.228713e-06 elapsed=08:07:17 avg_step=1.33s recent_step=1.23s eta=28:45:38 +2026-08-25 12:02:59 | INFO | train rank=0 | step=22030/100000 progress=22.0% loss=0.064514 lr=9.227830e-06 elapsed=08:07:29 avg_step=1.33s recent_step=1.19s eta=28:45:20 +2026-08-25 12:03:14 | INFO | train rank=0 | step=22040/100000 progress=22.0% loss=0.064148 lr=9.226947e-06 elapsed=08:07:44 avg_step=1.33s recent_step=1.49s eta=28:45:13 +2026-08-25 12:03:26 | INFO | train rank=0 | step=22050/100000 progress=22.1% loss=0.069427 lr=9.226064e-06 elapsed=08:07:55 avg_step=1.33s recent_step=1.19s eta=28:44:54 +2026-08-25 12:03:40 | INFO | train rank=0 | step=22060/100000 progress=22.1% loss=0.052094 lr=9.225180e-06 elapsed=08:08:10 avg_step=1.33s recent_step=1.41s eta=28:44:44 +2026-08-25 12:03:51 | INFO | train rank=0 | step=22070/100000 progress=22.1% loss=0.061203 lr=9.224295e-06 elapsed=08:08:21 avg_step=1.33s recent_step=1.17s eta=28:44:25 +2026-08-25 12:04:05 | INFO | train rank=0 | step=22080/100000 progress=22.1% loss=0.041183 lr=9.223411e-06 elapsed=08:08:35 avg_step=1.33s recent_step=1.39s eta=28:44:14 +2026-08-25 12:04:18 | INFO | train rank=0 | step=22090/100000 progress=22.1% loss=0.093109 lr=9.222525e-06 elapsed=08:08:48 avg_step=1.33s recent_step=1.32s eta=28:44:01 +2026-08-25 12:04:34 | INFO | train rank=0 | step=22100/100000 progress=22.1% loss=0.057755 lr=9.221640e-06 elapsed=08:09:04 avg_step=1.33s recent_step=1.59s eta=28:43:56 +2026-08-25 12:04:46 | INFO | train rank=0 | step=22110/100000 progress=22.1% loss=0.039818 lr=9.220753e-06 elapsed=08:09:16 avg_step=1.33s recent_step=1.21s eta=28:43:39 +2026-08-25 12:04:58 | INFO | train rank=0 | step=22120/100000 progress=22.1% loss=0.062103 lr=9.219867e-06 elapsed=08:09:28 avg_step=1.33s recent_step=1.16s eta=28:43:20 +2026-08-25 12:05:10 | INFO | train rank=0 | step=22130/100000 progress=22.1% loss=0.056580 lr=9.218980e-06 elapsed=08:09:39 avg_step=1.33s recent_step=1.15s eta=28:43:00 +2026-08-25 12:05:23 | INFO | train rank=0 | step=22140/100000 progress=22.1% loss=0.127258 lr=9.218092e-06 elapsed=08:09:53 avg_step=1.33s recent_step=1.37s eta=28:42:49 +2026-08-25 12:05:37 | INFO | train rank=0 | step=22150/100000 progress=22.1% loss=0.092682 lr=9.217204e-06 elapsed=08:10:07 avg_step=1.33s recent_step=1.40s eta=28:42:38 +2026-08-25 12:05:48 | INFO | train rank=0 | step=22160/100000 progress=22.2% loss=0.092896 lr=9.216315e-06 elapsed=08:10:18 avg_step=1.33s recent_step=1.12s eta=28:42:17 +2026-08-25 12:06:00 | INFO | train rank=0 | step=22170/100000 progress=22.2% loss=0.053192 lr=9.215427e-06 elapsed=08:10:30 avg_step=1.33s recent_step=1.15s eta=28:41:58 +2026-08-25 12:06:15 | INFO | train rank=0 | step=22180/100000 progress=22.2% loss=0.049179 lr=9.214537e-06 elapsed=08:10:45 avg_step=1.33s recent_step=1.50s eta=28:41:50 +2026-08-25 12:06:30 | INFO | train rank=0 | step=22190/100000 progress=22.2% loss=0.052002 lr=9.213647e-06 elapsed=08:11:00 avg_step=1.33s recent_step=1.52s eta=28:41:44 +2026-08-25 12:06:43 | INFO | train rank=0 | step=22200/100000 progress=22.2% loss=0.045746 lr=9.212757e-06 elapsed=08:11:13 avg_step=1.33s recent_step=1.27s eta=28:41:29 +2026-08-25 12:06:54 | INFO | train rank=0 | step=22210/100000 progress=22.2% loss=0.056839 lr=9.211866e-06 elapsed=08:11:24 avg_step=1.33s recent_step=1.12s eta=28:41:08 +2026-08-25 12:07:13 | INFO | train rank=0 | step=22220/100000 progress=22.2% loss=0.053055 lr=9.210975e-06 elapsed=08:11:43 avg_step=1.33s recent_step=1.91s eta=28:41:15 +2026-08-25 12:07:28 | INFO | train rank=0 | step=22230/100000 progress=22.2% loss=0.085617 lr=9.210083e-06 elapsed=08:11:58 avg_step=1.33s recent_step=1.50s eta=28:41:08 +2026-08-25 12:07:41 | INFO | train rank=0 | step=22240/100000 progress=22.2% loss=0.061081 lr=9.209191e-06 elapsed=08:12:11 avg_step=1.33s recent_step=1.28s eta=28:40:53 +2026-08-25 12:07:55 | INFO | train rank=0 | step=22250/100000 progress=22.2% loss=0.063995 lr=9.208298e-06 elapsed=08:12:25 avg_step=1.33s recent_step=1.40s eta=28:40:43 +2026-08-25 12:08:06 | INFO | train rank=0 | step=22260/100000 progress=22.3% loss=0.100250 lr=9.207405e-06 elapsed=08:12:36 avg_step=1.33s recent_step=1.11s eta=28:40:22 +2026-08-25 12:08:21 | INFO | train rank=0 | step=22270/100000 progress=22.3% loss=0.066772 lr=9.206511e-06 elapsed=08:12:51 avg_step=1.33s recent_step=1.49s eta=28:40:14 +2026-08-25 12:08:35 | INFO | train rank=0 | step=22280/100000 progress=22.3% loss=0.039276 lr=9.205617e-06 elapsed=08:13:05 avg_step=1.33s recent_step=1.40s eta=28:40:04 +2026-08-25 12:08:48 | INFO | train rank=0 | step=22290/100000 progress=22.3% loss=0.043732 lr=9.204723e-06 elapsed=08:13:18 avg_step=1.33s recent_step=1.30s eta=28:39:49 +2026-08-25 12:09:00 | INFO | train rank=0 | step=22300/100000 progress=22.3% loss=0.070648 lr=9.203828e-06 elapsed=08:13:29 avg_step=1.33s recent_step=1.15s eta=28:39:30 +2026-08-25 12:09:11 | INFO | train rank=0 | step=22310/100000 progress=22.3% loss=0.051239 lr=9.202933e-06 elapsed=08:13:41 avg_step=1.33s recent_step=1.13s eta=28:39:09 +2026-08-25 12:09:30 | INFO | train rank=0 | step=22320/100000 progress=22.3% loss=0.070648 lr=9.202037e-06 elapsed=08:14:00 avg_step=1.33s recent_step=1.95s eta=28:39:18 +2026-08-25 12:09:45 | INFO | train rank=0 | step=22330/100000 progress=22.3% loss=0.075012 lr=9.201140e-06 elapsed=08:14:15 avg_step=1.33s recent_step=1.50s eta=28:39:11 +2026-08-25 12:09:56 | INFO | train rank=0 | step=22340/100000 progress=22.3% loss=0.075623 lr=9.200244e-06 elapsed=08:14:26 avg_step=1.33s recent_step=1.11s eta=28:38:50 +2026-08-25 12:10:09 | INFO | train rank=0 | step=22350/100000 progress=22.4% loss=0.097046 lr=9.199346e-06 elapsed=08:14:38 avg_step=1.33s recent_step=1.21s eta=28:38:32 +2026-08-25 12:10:22 | INFO | train rank=0 | step=22360/100000 progress=22.4% loss=0.062988 lr=9.198449e-06 elapsed=08:14:52 avg_step=1.33s recent_step=1.32s eta=28:38:19 +2026-08-25 12:10:33 | INFO | train rank=0 | step=22370/100000 progress=22.4% loss=0.098267 lr=9.197550e-06 elapsed=08:15:03 avg_step=1.33s recent_step=1.15s eta=28:37:59 +2026-08-25 12:10:49 | INFO | train rank=0 | step=22380/100000 progress=22.4% loss=0.080627 lr=9.196652e-06 elapsed=08:15:19 avg_step=1.33s recent_step=1.59s eta=28:37:55 +2026-08-25 12:11:07 | INFO | train rank=0 | step=22390/100000 progress=22.4% loss=0.074188 lr=9.195753e-06 elapsed=08:15:37 avg_step=1.33s recent_step=1.75s eta=28:37:56 +2026-08-25 12:11:18 | INFO | train rank=0 | step=22400/100000 progress=22.4% loss=0.061890 lr=9.194853e-06 elapsed=08:15:48 avg_step=1.33s recent_step=1.14s eta=28:37:36 +2026-08-25 12:11:30 | INFO | train rank=0 | step=22410/100000 progress=22.4% loss=0.082916 lr=9.193953e-06 elapsed=08:16:00 avg_step=1.33s recent_step=1.18s eta=28:37:18 +2026-08-25 12:11:42 | INFO | train rank=0 | step=22420/100000 progress=22.4% loss=0.100159 lr=9.193053e-06 elapsed=08:16:12 avg_step=1.33s recent_step=1.18s eta=28:37:00 +2026-08-25 12:11:57 | INFO | train rank=0 | step=22430/100000 progress=22.4% loss=0.055557 lr=9.192152e-06 elapsed=08:16:27 avg_step=1.33s recent_step=1.53s eta=28:36:54 +2026-08-25 12:12:15 | INFO | train rank=0 | step=22440/100000 progress=22.4% loss=0.104126 lr=9.191250e-06 elapsed=08:16:44 avg_step=1.33s recent_step=1.77s eta=28:36:55 +2026-08-25 12:12:26 | INFO | train rank=0 | step=22450/100000 progress=22.4% loss=0.100067 lr=9.190348e-06 elapsed=08:16:56 avg_step=1.33s recent_step=1.13s eta=28:36:35 +2026-08-25 12:12:39 | INFO | train rank=0 | step=22460/100000 progress=22.5% loss=0.074860 lr=9.189446e-06 elapsed=08:17:08 avg_step=1.33s recent_step=1.26s eta=28:36:20 +2026-08-25 12:12:53 | INFO | train rank=0 | step=22470/100000 progress=22.5% loss=0.051025 lr=9.188543e-06 elapsed=08:17:23 avg_step=1.33s recent_step=1.42s eta=28:36:09 +2026-08-25 12:13:07 | INFO | train rank=0 | step=22480/100000 progress=22.5% loss=0.053909 lr=9.187640e-06 elapsed=08:17:37 avg_step=1.33s recent_step=1.46s eta=28:36:01 +2026-08-25 12:13:19 | INFO | train rank=0 | step=22490/100000 progress=22.5% loss=0.079407 lr=9.186737e-06 elapsed=08:17:49 avg_step=1.33s recent_step=1.21s eta=28:35:43 +2026-08-25 12:13:31 | INFO | train rank=0 | step=22500/100000 progress=22.5% loss=0.045654 lr=9.185832e-06 elapsed=08:18:01 avg_step=1.33s recent_step=1.14s eta=28:35:24 +2026-08-25 12:13:47 | INFO | train rank=0 | step=22510/100000 progress=22.5% loss=0.079620 lr=9.184928e-06 elapsed=08:18:17 avg_step=1.33s recent_step=1.66s eta=28:35:22 +2026-08-25 12:14:01 | INFO | train rank=0 | step=22520/100000 progress=22.5% loss=0.066589 lr=9.184023e-06 elapsed=08:18:31 avg_step=1.33s recent_step=1.35s eta=28:35:09 +2026-08-25 12:14:13 | INFO | train rank=0 | step=22530/100000 progress=22.5% loss=0.051804 lr=9.183117e-06 elapsed=08:18:43 avg_step=1.33s recent_step=1.17s eta=28:34:51 +2026-08-25 12:14:31 | INFO | train rank=0 | step=22540/100000 progress=22.5% loss=0.049423 lr=9.182211e-06 elapsed=08:19:01 avg_step=1.33s recent_step=1.82s eta=28:34:54 +2026-08-25 12:14:42 | INFO | train rank=0 | step=22550/100000 progress=22.6% loss=0.068176 lr=9.181305e-06 elapsed=08:19:12 avg_step=1.33s recent_step=1.11s eta=28:34:34 +2026-08-25 12:14:54 | INFO | train rank=0 | step=22560/100000 progress=22.6% loss=0.081329 lr=9.180398e-06 elapsed=08:19:24 avg_step=1.33s recent_step=1.21s eta=28:34:17 +2026-08-25 12:15:07 | INFO | train rank=0 | step=22570/100000 progress=22.6% loss=0.132507 lr=9.179491e-06 elapsed=08:19:37 avg_step=1.33s recent_step=1.29s eta=28:34:02 +2026-08-25 12:15:19 | INFO | train rank=0 | step=22580/100000 progress=22.6% loss=0.060471 lr=9.178583e-06 elapsed=08:19:48 avg_step=1.33s recent_step=1.15s eta=28:33:43 +2026-08-25 12:15:33 | INFO | train rank=0 | step=22590/100000 progress=22.6% loss=0.064148 lr=9.177675e-06 elapsed=08:20:03 avg_step=1.33s recent_step=1.41s eta=28:33:32 +2026-08-25 12:15:50 | INFO | train rank=0 | step=22600/100000 progress=22.6% loss=0.032043 lr=9.176766e-06 elapsed=08:20:20 avg_step=1.33s recent_step=1.78s eta=28:33:34 +2026-08-25 12:16:02 | INFO | train rank=0 | step=22610/100000 progress=22.6% loss=0.069489 lr=9.175857e-06 elapsed=08:20:32 avg_step=1.33s recent_step=1.13s eta=28:33:14 +2026-08-25 12:16:14 | INFO | train rank=0 | step=22620/100000 progress=22.6% loss=0.052368 lr=9.174947e-06 elapsed=08:20:44 avg_step=1.33s recent_step=1.18s eta=28:32:56 +2026-08-25 12:16:29 | INFO | train rank=0 | step=22630/100000 progress=22.6% loss=0.087952 lr=9.174037e-06 elapsed=08:20:59 avg_step=1.33s recent_step=1.52s eta=28:32:49 +2026-08-25 12:16:41 | INFO | train rank=0 | step=22640/100000 progress=22.6% loss=0.053589 lr=9.173127e-06 elapsed=08:21:11 avg_step=1.33s recent_step=1.18s eta=28:32:31 +2026-08-25 12:16:53 | INFO | train rank=0 | step=22650/100000 progress=22.7% loss=0.065887 lr=9.172216e-06 elapsed=08:21:23 avg_step=1.33s recent_step=1.26s eta=28:32:15 +2026-08-25 12:17:07 | INFO | train rank=0 | step=22660/100000 progress=22.7% loss=0.035156 lr=9.171304e-06 elapsed=08:21:37 avg_step=1.33s recent_step=1.42s eta=28:32:05 +2026-08-25 12:17:19 | INFO | train rank=0 | step=22670/100000 progress=22.7% loss=0.067612 lr=9.170392e-06 elapsed=08:21:49 avg_step=1.33s recent_step=1.16s eta=28:31:46 +2026-08-25 12:17:34 | INFO | train rank=0 | step=22680/100000 progress=22.7% loss=0.092102 lr=9.169480e-06 elapsed=08:22:04 avg_step=1.33s recent_step=1.49s eta=28:31:38 +2026-08-25 12:17:50 | INFO | train rank=0 | step=22690/100000 progress=22.7% loss=0.072205 lr=9.168567e-06 elapsed=08:22:20 avg_step=1.33s recent_step=1.59s eta=28:31:34 +2026-08-25 12:18:01 | INFO | train rank=0 | step=22700/100000 progress=22.7% loss=0.053467 lr=9.167654e-06 elapsed=08:22:31 avg_step=1.33s recent_step=1.14s eta=28:31:14 +2026-08-25 12:18:13 | INFO | train rank=0 | step=22710/100000 progress=22.7% loss=0.064163 lr=9.166740e-06 elapsed=08:22:43 avg_step=1.33s recent_step=1.15s eta=28:30:55 +2026-08-25 12:18:26 | INFO | train rank=0 | step=22720/100000 progress=22.7% loss=0.091705 lr=9.165826e-06 elapsed=08:22:56 avg_step=1.33s recent_step=1.34s eta=28:30:42 +2026-08-25 12:18:42 | INFO | train rank=0 | step=22730/100000 progress=22.7% loss=0.060196 lr=9.164911e-06 elapsed=08:23:12 avg_step=1.33s recent_step=1.56s eta=28:30:37 +2026-08-25 12:18:59 | INFO | train rank=0 | step=22740/100000 progress=22.7% loss=0.052170 lr=9.163996e-06 elapsed=08:23:29 avg_step=1.33s recent_step=1.74s eta=28:30:37 +2026-08-25 12:19:11 | INFO | train rank=0 | step=22750/100000 progress=22.8% loss=0.059189 lr=9.163081e-06 elapsed=08:23:41 avg_step=1.33s recent_step=1.17s eta=28:30:18 +2026-08-25 12:19:22 | INFO | train rank=0 | step=22760/100000 progress=22.8% loss=0.049179 lr=9.162165e-06 elapsed=08:23:52 avg_step=1.33s recent_step=1.13s eta=28:29:58 +2026-08-25 12:19:35 | INFO | train rank=0 | step=22770/100000 progress=22.8% loss=0.087830 lr=9.161248e-06 elapsed=08:24:05 avg_step=1.33s recent_step=1.29s eta=28:29:44 +2026-08-25 12:19:47 | INFO | train rank=0 | step=22780/100000 progress=22.8% loss=0.092499 lr=9.160331e-06 elapsed=08:24:17 avg_step=1.33s recent_step=1.23s eta=28:29:27 +2026-08-25 12:20:00 | INFO | train rank=0 | step=22790/100000 progress=22.8% loss=0.052505 lr=9.159414e-06 elapsed=08:24:30 avg_step=1.33s recent_step=1.25s eta=28:29:11 +2026-08-25 12:20:14 | INFO | train rank=0 | step=22800/100000 progress=22.8% loss=0.130798 lr=9.158496e-06 elapsed=08:24:44 avg_step=1.33s recent_step=1.46s eta=28:29:03 +2026-08-25 12:20:26 | INFO | train rank=0 | step=22810/100000 progress=22.8% loss=0.055908 lr=9.157578e-06 elapsed=08:24:56 avg_step=1.33s recent_step=1.15s eta=28:28:43 +2026-08-25 12:20:39 | INFO | train rank=0 | step=22820/100000 progress=22.8% loss=0.129272 lr=9.156659e-06 elapsed=08:25:08 avg_step=1.33s recent_step=1.27s eta=28:28:28 +2026-08-25 12:20:51 | INFO | train rank=0 | step=22830/100000 progress=22.8% loss=0.045059 lr=9.155740e-06 elapsed=08:25:20 avg_step=1.33s recent_step=1.20s eta=28:28:10 +2026-08-25 12:21:04 | INFO | train rank=0 | step=22840/100000 progress=22.8% loss=0.061096 lr=9.154820e-06 elapsed=08:25:34 avg_step=1.33s recent_step=1.34s eta=28:27:57 +2026-08-25 12:21:16 | INFO | train rank=0 | step=22850/100000 progress=22.9% loss=0.039948 lr=9.153900e-06 elapsed=08:25:46 avg_step=1.33s recent_step=1.18s eta=28:27:39 +2026-08-25 12:21:27 | INFO | train rank=0 | step=22860/100000 progress=22.9% loss=0.067139 lr=9.152979e-06 elapsed=08:25:57 avg_step=1.33s recent_step=1.18s eta=28:27:21 +2026-08-25 12:21:41 | INFO | train rank=0 | step=22870/100000 progress=22.9% loss=0.081085 lr=9.152058e-06 elapsed=08:26:11 avg_step=1.33s recent_step=1.35s eta=28:27:08 +2026-08-25 12:21:55 | INFO | train rank=0 | step=22880/100000 progress=22.9% loss=0.048859 lr=9.151137e-06 elapsed=08:26:25 avg_step=1.33s recent_step=1.43s eta=28:26:58 +2026-08-25 12:22:09 | INFO | train rank=0 | step=22890/100000 progress=22.9% loss=0.072052 lr=9.150215e-06 elapsed=08:26:39 avg_step=1.33s recent_step=1.37s eta=28:26:46 +2026-08-25 12:22:21 | INFO | train rank=0 | step=22900/100000 progress=22.9% loss=0.068863 lr=9.149293e-06 elapsed=08:26:51 avg_step=1.33s recent_step=1.21s eta=28:26:29 +2026-08-25 12:22:38 | INFO | train rank=0 | step=22910/100000 progress=22.9% loss=0.056870 lr=9.148370e-06 elapsed=08:27:08 avg_step=1.33s recent_step=1.73s eta=28:26:29 +2026-08-25 12:22:50 | INFO | train rank=0 | step=22920/100000 progress=22.9% loss=0.048370 lr=9.147447e-06 elapsed=08:27:20 avg_step=1.33s recent_step=1.21s eta=28:26:12 +2026-08-25 12:23:03 | INFO | train rank=0 | step=22930/100000 progress=22.9% loss=0.051819 lr=9.146523e-06 elapsed=08:27:32 avg_step=1.33s recent_step=1.21s eta=28:25:55 +2026-08-25 12:23:19 | INFO | train rank=0 | step=22940/100000 progress=22.9% loss=0.058304 lr=9.145599e-06 elapsed=08:27:49 avg_step=1.33s recent_step=1.65s eta=28:25:52 +2026-08-25 12:23:32 | INFO | train rank=0 | step=22950/100000 progress=22.9% loss=0.042816 lr=9.144674e-06 elapsed=08:28:02 avg_step=1.33s recent_step=1.31s eta=28:25:38 +2026-08-25 12:23:50 | INFO | train rank=0 | step=22960/100000 progress=23.0% loss=0.098999 lr=9.143749e-06 elapsed=08:28:20 avg_step=1.33s recent_step=1.79s eta=28:25:41 +2026-08-25 12:24:01 | INFO | train rank=0 | step=22970/100000 progress=23.0% loss=0.117371 lr=9.142823e-06 elapsed=08:28:31 avg_step=1.33s recent_step=1.15s eta=28:25:21 +2026-08-25 12:24:13 | INFO | train rank=0 | step=22980/100000 progress=23.0% loss=0.087280 lr=9.141898e-06 elapsed=08:28:43 avg_step=1.33s recent_step=1.15s eta=28:25:02 +2026-08-25 12:24:25 | INFO | train rank=0 | step=22990/100000 progress=23.0% loss=0.051849 lr=9.140971e-06 elapsed=08:28:55 avg_step=1.33s recent_step=1.17s eta=28:24:43 +2026-08-25 12:24:40 | INFO | train rank=0 | step=23000/100000 progress=23.0% loss=0.044571 lr=9.140044e-06 elapsed=08:29:10 avg_step=1.33s recent_step=1.52s eta=28:24:37 +2026-08-25 12:24:54 | INFO | train rank=0 | step=23010/100000 progress=23.0% loss=0.042999 lr=9.139117e-06 elapsed=08:29:23 avg_step=1.33s recent_step=1.37s eta=28:24:25 +2026-08-25 12:25:06 | INFO | train rank=0 | step=23020/100000 progress=23.0% loss=0.047882 lr=9.138189e-06 elapsed=08:29:36 avg_step=1.33s recent_step=1.25s eta=28:24:09 +2026-08-25 12:25:22 | INFO | train rank=0 | step=23030/100000 progress=23.0% loss=0.063141 lr=9.137261e-06 elapsed=08:29:51 avg_step=1.33s recent_step=1.55s eta=28:24:03 +2026-08-25 12:25:35 | INFO | train rank=0 | step=23040/100000 progress=23.0% loss=0.076813 lr=9.136332e-06 elapsed=08:30:04 avg_step=1.33s recent_step=1.30s eta=28:23:49 +2026-08-25 12:25:49 | INFO | train rank=0 | step=23050/100000 progress=23.1% loss=0.075790 lr=9.135403e-06 elapsed=08:30:18 avg_step=1.33s recent_step=1.40s eta=28:23:38 +2026-08-25 12:26:01 | INFO | train rank=0 | step=23060/100000 progress=23.1% loss=0.060257 lr=9.134473e-06 elapsed=08:30:31 avg_step=1.33s recent_step=1.28s eta=28:23:23 +2026-08-25 12:26:13 | INFO | train rank=0 | step=23070/100000 progress=23.1% loss=0.085327 lr=9.133543e-06 elapsed=08:30:43 avg_step=1.33s recent_step=1.20s eta=28:23:05 +2026-08-25 12:26:30 | INFO | train rank=0 | step=23080/100000 progress=23.1% loss=0.098755 lr=9.132613e-06 elapsed=08:31:00 avg_step=1.33s recent_step=1.70s eta=28:23:04 +2026-08-25 12:26:42 | INFO | train rank=0 | step=23090/100000 progress=23.1% loss=0.061676 lr=9.131682e-06 elapsed=08:31:12 avg_step=1.33s recent_step=1.18s eta=28:22:46 +2026-08-25 12:26:54 | INFO | train rank=0 | step=23100/100000 progress=23.1% loss=0.059082 lr=9.130750e-06 elapsed=08:31:24 avg_step=1.33s recent_step=1.16s eta=28:22:27 +2026-08-25 12:27:07 | INFO | train rank=0 | step=23110/100000 progress=23.1% loss=0.048569 lr=9.129818e-06 elapsed=08:31:37 avg_step=1.33s recent_step=1.33s eta=28:22:14 +2026-08-25 12:27:23 | INFO | train rank=0 | step=23120/100000 progress=23.1% loss=0.060333 lr=9.128886e-06 elapsed=08:31:53 avg_step=1.33s recent_step=1.58s eta=28:22:09 +2026-08-25 12:27:34 | INFO | train rank=0 | step=23130/100000 progress=23.1% loss=0.101501 lr=9.127953e-06 elapsed=08:32:04 avg_step=1.33s recent_step=1.12s eta=28:21:49 +2026-08-25 12:27:49 | INFO | train rank=0 | step=23140/100000 progress=23.1% loss=0.056534 lr=9.127020e-06 elapsed=08:32:19 avg_step=1.33s recent_step=1.54s eta=28:21:43 +2026-08-25 12:28:02 | INFO | train rank=0 | step=23150/100000 progress=23.2% loss=0.050995 lr=9.126086e-06 elapsed=08:32:32 avg_step=1.33s recent_step=1.30s eta=28:21:28 +2026-08-25 12:28:16 | INFO | train rank=0 | step=23160/100000 progress=23.2% loss=0.058197 lr=9.125152e-06 elapsed=08:32:45 avg_step=1.33s recent_step=1.30s eta=28:21:14 +2026-08-25 12:28:28 | INFO | train rank=0 | step=23170/100000 progress=23.2% loss=0.068176 lr=9.124218e-06 elapsed=08:32:57 avg_step=1.33s recent_step=1.20s eta=28:20:57 +2026-08-25 12:28:39 | INFO | train rank=0 | step=23180/100000 progress=23.2% loss=0.072388 lr=9.123283e-06 elapsed=08:33:09 avg_step=1.33s recent_step=1.19s eta=28:20:39 +2026-08-25 12:28:54 | INFO | train rank=0 | step=23190/100000 progress=23.2% loss=0.065491 lr=9.122347e-06 elapsed=08:33:24 avg_step=1.33s recent_step=1.46s eta=28:20:30 +2026-08-25 12:29:06 | INFO | train rank=0 | step=23200/100000 progress=23.2% loss=0.087311 lr=9.121411e-06 elapsed=08:33:35 avg_step=1.33s recent_step=1.15s eta=28:20:11 +2026-08-25 12:29:17 | INFO | train rank=0 | step=23210/100000 progress=23.2% loss=0.043747 lr=9.120475e-06 elapsed=08:33:47 avg_step=1.33s recent_step=1.17s eta=28:19:52 +2026-08-25 12:29:29 | INFO | train rank=0 | step=23220/100000 progress=23.2% loss=0.061737 lr=9.119538e-06 elapsed=08:33:59 avg_step=1.33s recent_step=1.21s eta=28:19:35 +2026-08-25 12:29:43 | INFO | train rank=0 | step=23230/100000 progress=23.2% loss=0.077789 lr=9.118601e-06 elapsed=08:34:13 avg_step=1.33s recent_step=1.34s eta=28:19:22 +2026-08-25 12:29:55 | INFO | train rank=0 | step=23240/100000 progress=23.2% loss=0.050354 lr=9.117663e-06 elapsed=08:34:25 avg_step=1.33s recent_step=1.27s eta=28:19:07 +2026-08-25 12:30:09 | INFO | train rank=0 | step=23250/100000 progress=23.2% loss=0.067200 lr=9.116725e-06 elapsed=08:34:39 avg_step=1.33s recent_step=1.38s eta=28:18:55 +2026-08-25 12:30:21 | INFO | train rank=0 | step=23260/100000 progress=23.3% loss=0.054123 lr=9.115786e-06 elapsed=08:34:51 avg_step=1.33s recent_step=1.16s eta=28:18:37 +2026-08-25 12:30:38 | INFO | train rank=0 | step=23270/100000 progress=23.3% loss=0.075623 lr=9.114847e-06 elapsed=08:35:08 avg_step=1.33s recent_step=1.70s eta=28:18:36 +2026-08-25 12:30:49 | INFO | train rank=0 | step=23280/100000 progress=23.3% loss=0.034103 lr=9.113908e-06 elapsed=08:35:19 avg_step=1.33s recent_step=1.16s eta=28:18:17 +2026-08-25 12:31:03 | INFO | train rank=0 | step=23290/100000 progress=23.3% loss=0.071625 lr=9.112968e-06 elapsed=08:35:33 avg_step=1.33s recent_step=1.37s eta=28:18:05 +2026-08-25 12:31:19 | INFO | train rank=0 | step=23300/100000 progress=23.3% loss=0.072479 lr=9.112027e-06 elapsed=08:35:48 avg_step=1.33s recent_step=1.53s eta=28:17:59 +2026-08-25 12:31:30 | INFO | train rank=0 | step=23310/100000 progress=23.3% loss=0.100830 lr=9.111086e-06 elapsed=08:36:00 avg_step=1.33s recent_step=1.18s eta=28:17:41 +2026-08-25 12:31:43 | INFO | train rank=0 | step=23320/100000 progress=23.3% loss=0.065887 lr=9.110145e-06 elapsed=08:36:12 avg_step=1.33s recent_step=1.22s eta=28:17:24 +2026-08-25 12:31:57 | INFO | train rank=0 | step=23330/100000 progress=23.3% loss=0.077576 lr=9.109203e-06 elapsed=08:36:27 avg_step=1.33s recent_step=1.49s eta=28:17:16 +2026-08-25 12:32:13 | INFO | train rank=0 | step=23340/100000 progress=23.3% loss=0.079224 lr=9.108261e-06 elapsed=08:36:42 avg_step=1.33s recent_step=1.52s eta=28:17:09 +2026-08-25 12:32:24 | INFO | train rank=0 | step=23350/100000 progress=23.4% loss=0.069519 lr=9.107318e-06 elapsed=08:36:54 avg_step=1.33s recent_step=1.14s eta=28:16:49 +2026-08-25 12:32:39 | INFO | train rank=0 | step=23360/100000 progress=23.4% loss=0.064484 lr=9.106375e-06 elapsed=08:37:09 avg_step=1.33s recent_step=1.50s eta=28:16:42 +2026-08-25 12:32:51 | INFO | train rank=0 | step=23370/100000 progress=23.4% loss=0.066681 lr=9.105431e-06 elapsed=08:37:21 avg_step=1.33s recent_step=1.16s eta=28:16:23 +2026-08-25 12:33:02 | INFO | train rank=0 | step=23380/100000 progress=23.4% loss=0.049408 lr=9.104487e-06 elapsed=08:37:32 avg_step=1.33s recent_step=1.18s eta=28:16:04 +2026-08-25 12:33:16 | INFO | train rank=0 | step=23390/100000 progress=23.4% loss=0.063843 lr=9.103543e-06 elapsed=08:37:46 avg_step=1.33s recent_step=1.35s eta=28:15:52 +2026-08-25 12:33:27 | INFO | train rank=0 | step=23400/100000 progress=23.4% loss=0.057159 lr=9.102598e-06 elapsed=08:37:57 avg_step=1.33s recent_step=1.14s eta=28:15:32 +2026-08-25 12:33:40 | INFO | train rank=0 | step=23410/100000 progress=23.4% loss=0.045349 lr=9.101653e-06 elapsed=08:38:09 avg_step=1.33s recent_step=1.22s eta=28:15:16 +2026-08-25 12:33:52 | INFO | train rank=0 | step=23420/100000 progress=23.4% loss=0.037277 lr=9.100707e-06 elapsed=08:38:22 avg_step=1.33s recent_step=1.29s eta=28:15:01 +2026-08-25 12:34:11 | INFO | train rank=0 | step=23430/100000 progress=23.4% loss=0.066956 lr=9.099761e-06 elapsed=08:38:41 avg_step=1.33s recent_step=1.83s eta=28:15:04 +2026-08-25 12:34:25 | INFO | train rank=0 | step=23440/100000 progress=23.4% loss=0.040054 lr=9.098814e-06 elapsed=08:38:55 avg_step=1.33s recent_step=1.47s eta=28:14:56 +2026-08-25 12:34:37 | INFO | train rank=0 | step=23450/100000 progress=23.4% loss=0.070862 lr=9.097867e-06 elapsed=08:39:07 avg_step=1.33s recent_step=1.13s eta=28:14:36 +2026-08-25 12:34:54 | INFO | train rank=0 | step=23460/100000 progress=23.5% loss=0.067963 lr=9.096919e-06 elapsed=08:39:24 avg_step=1.33s recent_step=1.76s eta=28:14:37 +2026-08-25 12:35:06 | INFO | train rank=0 | step=23470/100000 progress=23.5% loss=0.051834 lr=9.095971e-06 elapsed=08:39:36 avg_step=1.33s recent_step=1.14s eta=28:14:18 +2026-08-25 12:35:21 | INFO | train rank=0 | step=23480/100000 progress=23.5% loss=0.044098 lr=9.095022e-06 elapsed=08:39:51 avg_step=1.33s recent_step=1.57s eta=28:14:12 +2026-08-25 12:35:33 | INFO | train rank=0 | step=23490/100000 progress=23.5% loss=0.055161 lr=9.094073e-06 elapsed=08:40:03 avg_step=1.33s recent_step=1.15s eta=28:13:53 +2026-08-25 12:35:44 | INFO | train rank=0 | step=23500/100000 progress=23.5% loss=0.064224 lr=9.093124e-06 elapsed=08:40:14 avg_step=1.33s recent_step=1.09s eta=28:13:32 +2026-08-25 12:35:58 | INFO | train rank=0 | step=23510/100000 progress=23.5% loss=0.083069 lr=9.092174e-06 elapsed=08:40:28 avg_step=1.33s recent_step=1.42s eta=28:13:22 +2026-08-25 12:36:09 | INFO | train rank=0 | step=23520/100000 progress=23.5% loss=0.059326 lr=9.091224e-06 elapsed=08:40:39 avg_step=1.33s recent_step=1.11s eta=28:13:01 +2026-08-25 12:36:24 | INFO | train rank=0 | step=23530/100000 progress=23.5% loss=0.049332 lr=9.090273e-06 elapsed=08:40:54 avg_step=1.33s recent_step=1.47s eta=28:12:53 +2026-08-25 12:36:35 | INFO | train rank=0 | step=23540/100000 progress=23.5% loss=0.060394 lr=9.089322e-06 elapsed=08:41:05 avg_step=1.33s recent_step=1.15s eta=28:12:33 +2026-08-25 12:36:47 | INFO | train rank=0 | step=23550/100000 progress=23.5% loss=0.046799 lr=9.088370e-06 elapsed=08:41:17 avg_step=1.33s recent_step=1.15s eta=28:12:14 +2026-08-25 12:37:06 | INFO | train rank=0 | step=23560/100000 progress=23.6% loss=0.098572 lr=9.087418e-06 elapsed=08:41:36 avg_step=1.33s recent_step=1.91s eta=28:12:20 +2026-08-25 12:37:17 | INFO | train rank=0 | step=23570/100000 progress=23.6% loss=0.047852 lr=9.086466e-06 elapsed=08:41:47 avg_step=1.33s recent_step=1.14s eta=28:12:01 +2026-08-25 12:37:29 | INFO | train rank=0 | step=23580/100000 progress=23.6% loss=0.035416 lr=9.085513e-06 elapsed=08:41:59 avg_step=1.33s recent_step=1.17s eta=28:11:42 +2026-08-25 12:37:45 | INFO | train rank=0 | step=23590/100000 progress=23.6% loss=0.049438 lr=9.084559e-06 elapsed=08:42:15 avg_step=1.33s recent_step=1.60s eta=28:11:38 +2026-08-25 12:37:57 | INFO | train rank=0 | step=23600/100000 progress=23.6% loss=0.058578 lr=9.083605e-06 elapsed=08:42:27 avg_step=1.33s recent_step=1.17s eta=28:11:19 +2026-08-25 12:38:09 | INFO | train rank=0 | step=23610/100000 progress=23.6% loss=0.044708 lr=9.082651e-06 elapsed=08:42:39 avg_step=1.33s recent_step=1.26s eta=28:11:04 +2026-08-25 12:38:21 | INFO | train rank=0 | step=23620/100000 progress=23.6% loss=0.078857 lr=9.081696e-06 elapsed=08:42:51 avg_step=1.33s recent_step=1.13s eta=28:10:44 +2026-08-25 12:38:38 | INFO | train rank=0 | step=23630/100000 progress=23.6% loss=0.052185 lr=9.080741e-06 elapsed=08:43:08 avg_step=1.33s recent_step=1.78s eta=28:10:45 +2026-08-25 12:38:50 | INFO | train rank=0 | step=23640/100000 progress=23.6% loss=0.050888 lr=9.079785e-06 elapsed=08:43:20 avg_step=1.33s recent_step=1.17s eta=28:10:27 +2026-08-25 12:39:02 | INFO | train rank=0 | step=23650/100000 progress=23.6% loss=0.070015 lr=9.078829e-06 elapsed=08:43:32 avg_step=1.33s recent_step=1.19s eta=28:10:09 +2026-08-25 12:39:14 | INFO | train rank=0 | step=23660/100000 progress=23.7% loss=0.063934 lr=9.077873e-06 elapsed=08:43:44 avg_step=1.33s recent_step=1.19s eta=28:09:52 +2026-08-25 12:39:30 | INFO | train rank=0 | step=23670/100000 progress=23.7% loss=0.067810 lr=9.076916e-06 elapsed=08:44:00 avg_step=1.33s recent_step=1.62s eta=28:09:48 +2026-08-25 12:39:44 | INFO | train rank=0 | step=23680/100000 progress=23.7% loss=0.061310 lr=9.075958e-06 elapsed=08:44:13 avg_step=1.33s recent_step=1.34s eta=28:09:35 +2026-08-25 12:39:56 | INFO | train rank=0 | step=23690/100000 progress=23.7% loss=0.067322 lr=9.075000e-06 elapsed=08:44:26 avg_step=1.33s recent_step=1.22s eta=28:09:18 +2026-08-25 12:40:09 | INFO | train rank=0 | step=23700/100000 progress=23.7% loss=0.047104 lr=9.074042e-06 elapsed=08:44:39 avg_step=1.33s recent_step=1.37s eta=28:09:06 +2026-08-25 12:40:24 | INFO | train rank=0 | step=23710/100000 progress=23.7% loss=0.055740 lr=9.073083e-06 elapsed=08:44:54 avg_step=1.33s recent_step=1.49s eta=28:08:58 +2026-08-25 12:40:37 | INFO | train rank=0 | step=23720/100000 progress=23.7% loss=0.061523 lr=9.072124e-06 elapsed=08:45:06 avg_step=1.33s recent_step=1.22s eta=28:08:41 +2026-08-25 12:40:52 | INFO | train rank=0 | step=23730/100000 progress=23.7% loss=0.078217 lr=9.071164e-06 elapsed=08:45:22 avg_step=1.33s recent_step=1.58s eta=28:08:36 +2026-08-25 12:41:07 | INFO | train rank=0 | step=23740/100000 progress=23.7% loss=0.078247 lr=9.070204e-06 elapsed=08:45:37 avg_step=1.33s recent_step=1.51s eta=28:08:29 +2026-08-25 12:41:24 | INFO | train rank=0 | step=23750/100000 progress=23.8% loss=0.063255 lr=9.069244e-06 elapsed=08:45:54 avg_step=1.33s recent_step=1.67s eta=28:08:26 +2026-08-25 12:41:37 | INFO | train rank=0 | step=23760/100000 progress=23.8% loss=0.085663 lr=9.068283e-06 elapsed=08:46:07 avg_step=1.33s recent_step=1.31s eta=28:08:12 +2026-08-25 12:41:52 | INFO | train rank=0 | step=23770/100000 progress=23.8% loss=0.044319 lr=9.067321e-06 elapsed=08:46:22 avg_step=1.33s recent_step=1.46s eta=28:08:03 +2026-08-25 12:42:03 | INFO | train rank=0 | step=23780/100000 progress=23.8% loss=0.081116 lr=9.066359e-06 elapsed=08:46:33 avg_step=1.33s recent_step=1.12s eta=28:07:43 +2026-08-25 12:42:15 | INFO | train rank=0 | step=23790/100000 progress=23.8% loss=0.051514 lr=9.065397e-06 elapsed=08:46:45 avg_step=1.33s recent_step=1.17s eta=28:07:25 +2026-08-25 12:42:27 | INFO | train rank=0 | step=23800/100000 progress=23.8% loss=0.047836 lr=9.064434e-06 elapsed=08:46:57 avg_step=1.33s recent_step=1.20s eta=28:07:08 +2026-08-25 12:42:42 | INFO | train rank=0 | step=23810/100000 progress=23.8% loss=0.046387 lr=9.063471e-06 elapsed=08:47:12 avg_step=1.33s recent_step=1.51s eta=28:07:00 +2026-08-25 12:42:53 | INFO | train rank=0 | step=23820/100000 progress=23.8% loss=0.053726 lr=9.062507e-06 elapsed=08:47:23 avg_step=1.33s recent_step=1.13s eta=28:06:41 +2026-08-25 12:43:09 | INFO | train rank=0 | step=23830/100000 progress=23.8% loss=0.063721 lr=9.061543e-06 elapsed=08:47:38 avg_step=1.33s recent_step=1.52s eta=28:06:34 +2026-08-25 12:43:20 | INFO | train rank=0 | step=23840/100000 progress=23.8% loss=0.068329 lr=9.060578e-06 elapsed=08:47:50 avg_step=1.33s recent_step=1.15s eta=28:06:15 +2026-08-25 12:43:33 | INFO | train rank=0 | step=23850/100000 progress=23.8% loss=0.066162 lr=9.059613e-06 elapsed=08:48:02 avg_step=1.33s recent_step=1.26s eta=28:05:59 +2026-08-25 12:43:48 | INFO | train rank=0 | step=23860/100000 progress=23.9% loss=0.061493 lr=9.058648e-06 elapsed=08:48:18 avg_step=1.33s recent_step=1.57s eta=28:05:53 +2026-08-25 12:44:03 | INFO | train rank=0 | step=23870/100000 progress=23.9% loss=0.070099 lr=9.057682e-06 elapsed=08:48:33 avg_step=1.33s recent_step=1.47s eta=28:05:45 +2026-08-25 12:44:17 | INFO | train rank=0 | step=23880/100000 progress=23.9% loss=0.053375 lr=9.056716e-06 elapsed=08:48:46 avg_step=1.33s recent_step=1.37s eta=28:05:33 +2026-08-25 12:44:30 | INFO | train rank=0 | step=23890/100000 progress=23.9% loss=0.083923 lr=9.055749e-06 elapsed=08:49:00 avg_step=1.33s recent_step=1.31s eta=28:05:19 +2026-08-25 12:44:42 | INFO | train rank=0 | step=23900/100000 progress=23.9% loss=0.042755 lr=9.054782e-06 elapsed=08:49:11 avg_step=1.33s recent_step=1.19s eta=28:05:01 +2026-08-25 12:44:57 | INFO | train rank=0 | step=23910/100000 progress=23.9% loss=0.056656 lr=9.053814e-06 elapsed=08:49:27 avg_step=1.33s recent_step=1.54s eta=28:04:55 +2026-08-25 12:45:15 | INFO | train rank=0 | step=23920/100000 progress=23.9% loss=0.078827 lr=9.052846e-06 elapsed=08:49:45 avg_step=1.33s recent_step=1.76s eta=28:04:55 +2026-08-25 12:45:26 | INFO | train rank=0 | step=23930/100000 progress=23.9% loss=0.080109 lr=9.051877e-06 elapsed=08:49:56 avg_step=1.33s recent_step=1.12s eta=28:04:35 +2026-08-25 12:45:37 | INFO | train rank=0 | step=23940/100000 progress=23.9% loss=0.097595 lr=9.050908e-06 elapsed=08:50:07 avg_step=1.33s recent_step=1.14s eta=28:04:16 +2026-08-25 12:45:49 | INFO | train rank=0 | step=23950/100000 progress=23.9% loss=0.037155 lr=9.049939e-06 elapsed=08:50:19 avg_step=1.33s recent_step=1.21s eta=28:03:59 +2026-08-25 12:46:01 | INFO | train rank=0 | step=23960/100000 progress=24.0% loss=0.051819 lr=9.048969e-06 elapsed=08:50:31 avg_step=1.33s recent_step=1.15s eta=28:03:40 +2026-08-25 12:46:12 | INFO | train rank=0 | step=23970/100000 progress=24.0% loss=0.041733 lr=9.047999e-06 elapsed=08:50:42 avg_step=1.33s recent_step=1.14s eta=28:03:20 +2026-08-25 12:46:24 | INFO | train rank=0 | step=23980/100000 progress=24.0% loss=0.079102 lr=9.047028e-06 elapsed=08:50:54 avg_step=1.33s recent_step=1.17s eta=28:03:02 +2026-08-25 12:46:38 | INFO | train rank=0 | step=23990/100000 progress=24.0% loss=0.043762 lr=9.046057e-06 elapsed=08:51:08 avg_step=1.33s recent_step=1.44s eta=28:02:52 +2026-08-25 12:46:53 | INFO | train rank=0 | step=24000/100000 progress=24.0% loss=0.074142 lr=9.045085e-06 elapsed=08:51:23 avg_step=1.33s recent_step=1.50s eta=28:02:44 +2026-08-25 12:47:05 | INFO | train rank=0 | step=24010/100000 progress=24.0% loss=0.049698 lr=9.044113e-06 elapsed=08:51:35 avg_step=1.33s recent_step=1.22s eta=28:02:27 +2026-08-25 12:47:22 | INFO | train rank=0 | step=24020/100000 progress=24.0% loss=0.069580 lr=9.043140e-06 elapsed=08:51:52 avg_step=1.33s recent_step=1.67s eta=28:02:25 +2026-08-25 12:47:37 | INFO | train rank=0 | step=24030/100000 progress=24.0% loss=0.068237 lr=9.042167e-06 elapsed=08:52:07 avg_step=1.33s recent_step=1.50s eta=28:02:17 +2026-08-25 12:47:48 | INFO | train rank=0 | step=24040/100000 progress=24.0% loss=0.040707 lr=9.041194e-06 elapsed=08:52:18 avg_step=1.33s recent_step=1.13s eta=28:01:58 +2026-08-25 12:48:00 | INFO | train rank=0 | step=24050/100000 progress=24.1% loss=0.050461 lr=9.040220e-06 elapsed=08:52:30 avg_step=1.33s recent_step=1.21s eta=28:01:40 +2026-08-25 12:48:12 | INFO | train rank=0 | step=24060/100000 progress=24.1% loss=0.082733 lr=9.039246e-06 elapsed=08:52:41 avg_step=1.33s recent_step=1.10s eta=28:01:20 +2026-08-25 12:48:25 | INFO | train rank=0 | step=24070/100000 progress=24.1% loss=0.053955 lr=9.038271e-06 elapsed=08:52:55 avg_step=1.33s recent_step=1.40s eta=28:01:09 +2026-08-25 12:48:40 | INFO | train rank=0 | step=24080/100000 progress=24.1% loss=0.042297 lr=9.037296e-06 elapsed=08:53:10 avg_step=1.33s recent_step=1.44s eta=28:00:59 +2026-08-25 12:48:54 | INFO | train rank=0 | step=24090/100000 progress=24.1% loss=0.066193 lr=9.036320e-06 elapsed=08:53:24 avg_step=1.33s recent_step=1.46s eta=28:00:50 +2026-08-25 12:49:07 | INFO | train rank=0 | step=24100/100000 progress=24.1% loss=0.058456 lr=9.035344e-06 elapsed=08:53:37 avg_step=1.33s recent_step=1.22s eta=28:00:33 +2026-08-25 12:49:21 | INFO | train rank=0 | step=24110/100000 progress=24.1% loss=0.046387 lr=9.034367e-06 elapsed=08:53:51 avg_step=1.33s recent_step=1.48s eta=28:00:25 +2026-08-25 12:49:34 | INFO | train rank=0 | step=24120/100000 progress=24.1% loss=0.070374 lr=9.033391e-06 elapsed=08:54:04 avg_step=1.33s recent_step=1.30s eta=28:00:10 +2026-08-25 12:49:49 | INFO | train rank=0 | step=24130/100000 progress=24.1% loss=0.054291 lr=9.032413e-06 elapsed=08:54:19 avg_step=1.33s recent_step=1.46s eta=28:00:01 +2026-08-25 12:50:01 | INFO | train rank=0 | step=24140/100000 progress=24.1% loss=0.057404 lr=9.031435e-06 elapsed=08:54:31 avg_step=1.33s recent_step=1.17s eta=27:59:43 +2026-08-25 12:50:18 | INFO | train rank=0 | step=24150/100000 progress=24.1% loss=0.058304 lr=9.030457e-06 elapsed=08:54:48 avg_step=1.33s recent_step=1.73s eta=27:59:42 +2026-08-25 12:50:30 | INFO | train rank=0 | step=24160/100000 progress=24.2% loss=0.033493 lr=9.029478e-06 elapsed=08:55:00 avg_step=1.33s recent_step=1.21s eta=27:59:26 +2026-08-25 12:50:44 | INFO | train rank=0 | step=24170/100000 progress=24.2% loss=0.078491 lr=9.028499e-06 elapsed=08:55:14 avg_step=1.33s recent_step=1.41s eta=27:59:15 +2026-08-25 12:50:58 | INFO | train rank=0 | step=24180/100000 progress=24.2% loss=0.063995 lr=9.027519e-06 elapsed=08:55:28 avg_step=1.33s recent_step=1.40s eta=27:59:04 +2026-08-25 12:51:10 | INFO | train rank=0 | step=24190/100000 progress=24.2% loss=0.052734 lr=9.026539e-06 elapsed=08:55:40 avg_step=1.33s recent_step=1.16s eta=27:58:45 +2026-08-25 12:51:21 | INFO | train rank=0 | step=24200/100000 progress=24.2% loss=0.053528 lr=9.025559e-06 elapsed=08:55:51 avg_step=1.33s recent_step=1.16s eta=27:58:26 +2026-08-25 12:51:33 | INFO | train rank=0 | step=24210/100000 progress=24.2% loss=0.096375 lr=9.024578e-06 elapsed=08:56:03 avg_step=1.33s recent_step=1.18s eta=27:58:09 +2026-08-25 12:51:45 | INFO | train rank=0 | step=24220/100000 progress=24.2% loss=0.070450 lr=9.023597e-06 elapsed=08:56:15 avg_step=1.33s recent_step=1.20s eta=27:57:51 +2026-08-25 12:51:59 | INFO | train rank=0 | step=24230/100000 progress=24.2% loss=0.066711 lr=9.022615e-06 elapsed=08:56:29 avg_step=1.33s recent_step=1.35s eta=27:57:39 +2026-08-25 12:52:12 | INFO | train rank=0 | step=24240/100000 progress=24.2% loss=0.047997 lr=9.021633e-06 elapsed=08:56:42 avg_step=1.33s recent_step=1.35s eta=27:57:26 +2026-08-25 12:52:29 | INFO | train rank=0 | step=24250/100000 progress=24.2% loss=0.046227 lr=9.020650e-06 elapsed=08:56:59 avg_step=1.33s recent_step=1.66s eta=27:57:23 +2026-08-25 12:52:41 | INFO | train rank=0 | step=24260/100000 progress=24.3% loss=0.067780 lr=9.019667e-06 elapsed=08:57:11 avg_step=1.33s recent_step=1.23s eta=27:57:07 +2026-08-25 12:52:54 | INFO | train rank=0 | step=24270/100000 progress=24.3% loss=0.043320 lr=9.018683e-06 elapsed=08:57:24 avg_step=1.33s recent_step=1.27s eta=27:56:52 +2026-08-25 12:53:05 | INFO | train rank=0 | step=24280/100000 progress=24.3% loss=0.045753 lr=9.017699e-06 elapsed=08:57:35 avg_step=1.33s recent_step=1.12s eta=27:56:32 +2026-08-25 12:53:18 | INFO | train rank=0 | step=24290/100000 progress=24.3% loss=0.055466 lr=9.016715e-06 elapsed=08:57:47 avg_step=1.33s recent_step=1.25s eta=27:56:16 +2026-08-25 12:53:32 | INFO | train rank=0 | step=24300/100000 progress=24.3% loss=0.063568 lr=9.015730e-06 elapsed=08:58:02 avg_step=1.33s recent_step=1.46s eta=27:56:07 +2026-08-25 12:53:46 | INFO | train rank=0 | step=24310/100000 progress=24.3% loss=0.081787 lr=9.014745e-06 elapsed=08:58:16 avg_step=1.33s recent_step=1.35s eta=27:55:54 +2026-08-25 12:53:58 | INFO | train rank=0 | step=24320/100000 progress=24.3% loss=0.043396 lr=9.013759e-06 elapsed=08:58:27 avg_step=1.33s recent_step=1.18s eta=27:55:37 +2026-08-25 12:54:09 | INFO | train rank=0 | step=24330/100000 progress=24.3% loss=0.051620 lr=9.012773e-06 elapsed=08:58:39 avg_step=1.33s recent_step=1.18s eta=27:55:19 +2026-08-25 12:54:26 | INFO | train rank=0 | step=24340/100000 progress=24.3% loss=0.049591 lr=9.011786e-06 elapsed=08:58:55 avg_step=1.33s recent_step=1.63s eta=27:55:15 +2026-08-25 12:54:40 | INFO | train rank=0 | step=24350/100000 progress=24.3% loss=0.040894 lr=9.010799e-06 elapsed=08:59:10 avg_step=1.33s recent_step=1.46s eta=27:55:06 +2026-08-25 12:54:53 | INFO | train rank=0 | step=24360/100000 progress=24.4% loss=0.058716 lr=9.009811e-06 elapsed=08:59:23 avg_step=1.33s recent_step=1.28s eta=27:54:51 +2026-08-25 12:55:05 | INFO | train rank=0 | step=24370/100000 progress=24.4% loss=0.046448 lr=9.008823e-06 elapsed=08:59:35 avg_step=1.33s recent_step=1.21s eta=27:54:34 +2026-08-25 12:55:23 | INFO | train rank=0 | step=24380/100000 progress=24.4% loss=0.050110 lr=9.007835e-06 elapsed=08:59:52 avg_step=1.33s recent_step=1.75s eta=27:54:34 +2026-08-25 12:55:35 | INFO | train rank=0 | step=24390/100000 progress=24.4% loss=0.050568 lr=9.006846e-06 elapsed=09:00:05 avg_step=1.33s recent_step=1.21s eta=27:54:17 +2026-08-25 12:55:46 | INFO | train rank=0 | step=24400/100000 progress=24.4% loss=0.065186 lr=9.005857e-06 elapsed=09:00:16 avg_step=1.33s recent_step=1.15s eta=27:53:58 +2026-08-25 12:56:00 | INFO | train rank=0 | step=24410/100000 progress=24.4% loss=0.064835 lr=9.004867e-06 elapsed=09:00:30 avg_step=1.33s recent_step=1.37s eta=27:53:46 +2026-08-25 12:56:12 | INFO | train rank=0 | step=24420/100000 progress=24.4% loss=0.051315 lr=9.003877e-06 elapsed=09:00:42 avg_step=1.33s recent_step=1.20s eta=27:53:28 +2026-08-25 12:56:26 | INFO | train rank=0 | step=24430/100000 progress=24.4% loss=0.082489 lr=9.002886e-06 elapsed=09:00:56 avg_step=1.33s recent_step=1.44s eta=27:53:18 +2026-08-25 12:56:41 | INFO | train rank=0 | step=24440/100000 progress=24.4% loss=0.068268 lr=9.001895e-06 elapsed=09:01:11 avg_step=1.33s recent_step=1.52s eta=27:53:11 +2026-08-25 12:56:53 | INFO | train rank=0 | step=24450/100000 progress=24.4% loss=0.075714 lr=9.000904e-06 elapsed=09:01:23 avg_step=1.33s recent_step=1.20s eta=27:52:54 +2026-08-25 12:57:05 | INFO | train rank=0 | step=24460/100000 progress=24.5% loss=0.086304 lr=8.999912e-06 elapsed=09:01:35 avg_step=1.33s recent_step=1.15s eta=27:52:35 +2026-08-25 12:57:21 | INFO | train rank=0 | step=24470/100000 progress=24.5% loss=0.091858 lr=8.998920e-06 elapsed=09:01:51 avg_step=1.33s recent_step=1.61s eta=27:52:30 +2026-08-25 12:57:37 | INFO | train rank=0 | step=24480/100000 progress=24.5% loss=0.037712 lr=8.997927e-06 elapsed=09:02:07 avg_step=1.33s recent_step=1.58s eta=27:52:25 +2026-08-25 12:57:48 | INFO | train rank=0 | step=24490/100000 progress=24.5% loss=0.056183 lr=8.996934e-06 elapsed=09:02:18 avg_step=1.33s recent_step=1.12s eta=27:52:05 +2026-08-25 12:58:00 | INFO | train rank=0 | step=24500/100000 progress=24.5% loss=0.069458 lr=8.995940e-06 elapsed=09:02:30 avg_step=1.33s recent_step=1.19s eta=27:51:47 +2026-08-25 12:58:11 | INFO | train rank=0 | step=24510/100000 progress=24.5% loss=0.063995 lr=8.994946e-06 elapsed=09:02:41 avg_step=1.33s recent_step=1.14s eta=27:51:28 +2026-08-25 12:58:23 | INFO | train rank=0 | step=24520/100000 progress=24.5% loss=0.050987 lr=8.993951e-06 elapsed=09:02:53 avg_step=1.33s recent_step=1.21s eta=27:51:11 +2026-08-25 12:58:35 | INFO | train rank=0 | step=24530/100000 progress=24.5% loss=0.084961 lr=8.992956e-06 elapsed=09:03:05 avg_step=1.33s recent_step=1.15s eta=27:50:53 +2026-08-25 12:58:47 | INFO | train rank=0 | step=24540/100000 progress=24.5% loss=0.059143 lr=8.991961e-06 elapsed=09:03:16 avg_step=1.33s recent_step=1.17s eta=27:50:34 +2026-08-25 12:58:58 | INFO | train rank=0 | step=24550/100000 progress=24.6% loss=0.068512 lr=8.990965e-06 elapsed=09:03:28 avg_step=1.33s recent_step=1.18s eta=27:50:17 +2026-08-25 12:59:13 | INFO | train rank=0 | step=24560/100000 progress=24.6% loss=0.065750 lr=8.989969e-06 elapsed=09:03:43 avg_step=1.33s recent_step=1.49s eta=27:50:08 +2026-08-25 12:59:28 | INFO | train rank=0 | step=24570/100000 progress=24.6% loss=0.068329 lr=8.988972e-06 elapsed=09:03:58 avg_step=1.33s recent_step=1.47s eta=27:49:59 +2026-08-25 12:59:42 | INFO | train rank=0 | step=24580/100000 progress=24.6% loss=0.056015 lr=8.987975e-06 elapsed=09:04:12 avg_step=1.33s recent_step=1.45s eta=27:49:50 +2026-08-25 12:59:54 | INFO | train rank=0 | step=24590/100000 progress=24.6% loss=0.064209 lr=8.986977e-06 elapsed=09:04:23 avg_step=1.33s recent_step=1.12s eta=27:49:30 +2026-08-25 13:00:06 | INFO | train rank=0 | step=24600/100000 progress=24.6% loss=0.028763 lr=8.985979e-06 elapsed=09:04:36 avg_step=1.33s recent_step=1.21s eta=27:49:13 +2026-08-25 13:00:17 | INFO | train rank=0 | step=24610/100000 progress=24.6% loss=0.087097 lr=8.984981e-06 elapsed=09:04:47 avg_step=1.33s recent_step=1.18s eta=27:48:55 +2026-08-25 13:00:32 | INFO | train rank=0 | step=24620/100000 progress=24.6% loss=0.058075 lr=8.983982e-06 elapsed=09:05:02 avg_step=1.33s recent_step=1.44s eta=27:48:45 +2026-08-25 13:00:49 | INFO | train rank=0 | step=24630/100000 progress=24.6% loss=0.065247 lr=8.982983e-06 elapsed=09:05:19 avg_step=1.33s recent_step=1.71s eta=27:48:43 +2026-08-25 13:01:02 | INFO | train rank=0 | step=24640/100000 progress=24.6% loss=0.056458 lr=8.981983e-06 elapsed=09:05:31 avg_step=1.33s recent_step=1.27s eta=27:48:28 +2026-08-25 13:01:16 | INFO | train rank=0 | step=24650/100000 progress=24.6% loss=0.052200 lr=8.980983e-06 elapsed=09:05:46 avg_step=1.33s recent_step=1.48s eta=27:48:20 +2026-08-25 13:01:28 | INFO | train rank=0 | step=24660/100000 progress=24.7% loss=0.041107 lr=8.979982e-06 elapsed=09:05:58 avg_step=1.33s recent_step=1.20s eta=27:48:02 +2026-08-25 13:01:40 | INFO | train rank=0 | step=24670/100000 progress=24.7% loss=0.072235 lr=8.978981e-06 elapsed=09:06:10 avg_step=1.33s recent_step=1.19s eta=27:47:45 +2026-08-25 13:01:54 | INFO | train rank=0 | step=24680/100000 progress=24.7% loss=0.043060 lr=8.977980e-06 elapsed=09:06:24 avg_step=1.33s recent_step=1.40s eta=27:47:34 +2026-08-25 13:02:08 | INFO | train rank=0 | step=24690/100000 progress=24.7% loss=0.050262 lr=8.976978e-06 elapsed=09:06:38 avg_step=1.33s recent_step=1.40s eta=27:47:23 +2026-08-25 13:02:24 | INFO | train rank=0 | step=24700/100000 progress=24.7% loss=0.076874 lr=8.975975e-06 elapsed=09:06:54 avg_step=1.33s recent_step=1.62s eta=27:47:18 +2026-08-25 13:02:36 | INFO | train rank=0 | step=24710/100000 progress=24.7% loss=0.034393 lr=8.974973e-06 elapsed=09:07:06 avg_step=1.33s recent_step=1.16s eta=27:47:00 +2026-08-25 13:02:48 | INFO | train rank=0 | step=24720/100000 progress=24.7% loss=0.058899 lr=8.973969e-06 elapsed=09:07:18 avg_step=1.33s recent_step=1.19s eta=27:46:42 +2026-08-25 13:03:00 | INFO | train rank=0 | step=24730/100000 progress=24.7% loss=0.051926 lr=8.972966e-06 elapsed=09:07:30 avg_step=1.33s recent_step=1.18s eta=27:46:24 +2026-08-25 13:03:11 | INFO | train rank=0 | step=24740/100000 progress=24.7% loss=0.047348 lr=8.971962e-06 elapsed=09:07:41 avg_step=1.33s recent_step=1.17s eta=27:46:06 +2026-08-25 13:03:23 | INFO | train rank=0 | step=24750/100000 progress=24.8% loss=0.050308 lr=8.970957e-06 elapsed=09:07:53 avg_step=1.33s recent_step=1.17s eta=27:45:48 +2026-08-25 13:03:35 | INFO | train rank=0 | step=24760/100000 progress=24.8% loss=0.043732 lr=8.969952e-06 elapsed=09:08:05 avg_step=1.33s recent_step=1.23s eta=27:45:32 +2026-08-25 13:03:50 | INFO | train rank=0 | step=24770/100000 progress=24.8% loss=0.060043 lr=8.968947e-06 elapsed=09:08:19 avg_step=1.33s recent_step=1.42s eta=27:45:21 +2026-08-25 13:04:02 | INFO | train rank=0 | step=24780/100000 progress=24.8% loss=0.077698 lr=8.967941e-06 elapsed=09:08:32 avg_step=1.33s recent_step=1.23s eta=27:45:05 +2026-08-25 13:04:16 | INFO | train rank=0 | step=24790/100000 progress=24.8% loss=0.047272 lr=8.966934e-06 elapsed=09:08:46 avg_step=1.33s recent_step=1.46s eta=27:44:56 +2026-08-25 13:04:31 | INFO | train rank=0 | step=24800/100000 progress=24.8% loss=0.065491 lr=8.965928e-06 elapsed=09:09:01 avg_step=1.33s recent_step=1.50s eta=27:44:48 +2026-08-25 13:04:44 | INFO | train rank=0 | step=24810/100000 progress=24.8% loss=0.087494 lr=8.964921e-06 elapsed=09:09:14 avg_step=1.33s recent_step=1.29s eta=27:44:33 +2026-08-25 13:04:56 | INFO | train rank=0 | step=24820/100000 progress=24.8% loss=0.036972 lr=8.963913e-06 elapsed=09:09:26 avg_step=1.33s recent_step=1.15s eta=27:44:15 +2026-08-25 13:05:11 | INFO | train rank=0 | step=24830/100000 progress=24.8% loss=0.059204 lr=8.962905e-06 elapsed=09:09:41 avg_step=1.33s recent_step=1.56s eta=27:44:08 +2026-08-25 13:05:23 | INFO | train rank=0 | step=24840/100000 progress=24.8% loss=0.037827 lr=8.961897e-06 elapsed=09:09:53 avg_step=1.33s recent_step=1.19s eta=27:43:51 +2026-08-25 13:05:38 | INFO | train rank=0 | step=24850/100000 progress=24.9% loss=0.067230 lr=8.960888e-06 elapsed=09:10:08 avg_step=1.33s recent_step=1.49s eta=27:43:43 +2026-08-25 13:05:50 | INFO | train rank=0 | step=24860/100000 progress=24.9% loss=0.044846 lr=8.959878e-06 elapsed=09:10:20 avg_step=1.33s recent_step=1.17s eta=27:43:25 +2026-08-25 13:06:04 | INFO | train rank=0 | step=24870/100000 progress=24.9% loss=0.049042 lr=8.958869e-06 elapsed=09:10:34 avg_step=1.33s recent_step=1.37s eta=27:43:13 +2026-08-25 13:06:19 | INFO | train rank=0 | step=24880/100000 progress=24.9% loss=0.081543 lr=8.957858e-06 elapsed=09:10:49 avg_step=1.33s recent_step=1.50s eta=27:43:05 +2026-08-25 13:06:33 | INFO | train rank=0 | step=24890/100000 progress=24.9% loss=0.059357 lr=8.956848e-06 elapsed=09:11:03 avg_step=1.33s recent_step=1.42s eta=27:42:54 +2026-08-25 13:06:47 | INFO | train rank=0 | step=24900/100000 progress=24.9% loss=0.072983 lr=8.955837e-06 elapsed=09:11:17 avg_step=1.33s recent_step=1.43s eta=27:42:44 +2026-08-25 13:06:59 | INFO | train rank=0 | step=24910/100000 progress=24.9% loss=0.046402 lr=8.954825e-06 elapsed=09:11:29 avg_step=1.33s recent_step=1.16s eta=27:42:25 +2026-08-25 13:07:11 | INFO | train rank=0 | step=24920/100000 progress=24.9% loss=0.065277 lr=8.953813e-06 elapsed=09:11:41 avg_step=1.33s recent_step=1.23s eta=27:42:09 +2026-08-25 13:07:23 | INFO | train rank=0 | step=24930/100000 progress=24.9% loss=0.081207 lr=8.952801e-06 elapsed=09:11:52 avg_step=1.33s recent_step=1.15s eta=27:41:50 +2026-08-25 13:07:37 | INFO | train rank=0 | step=24940/100000 progress=24.9% loss=0.042969 lr=8.951788e-06 elapsed=09:12:07 avg_step=1.33s recent_step=1.48s eta=27:41:42 +2026-08-25 13:07:51 | INFO | train rank=0 | step=24950/100000 progress=24.9% loss=0.056686 lr=8.950775e-06 elapsed=09:12:21 avg_step=1.33s recent_step=1.34s eta=27:41:29 +2026-08-25 13:08:06 | INFO | train rank=0 | step=24960/100000 progress=25.0% loss=0.038849 lr=8.949761e-06 elapsed=09:12:36 avg_step=1.33s recent_step=1.49s eta=27:41:20 +2026-08-25 13:08:17 | INFO | train rank=0 | step=24970/100000 progress=25.0% loss=0.070007 lr=8.948747e-06 elapsed=09:12:47 avg_step=1.33s recent_step=1.14s eta=27:41:01 +2026-08-25 13:08:34 | INFO | train rank=0 | step=24980/100000 progress=25.0% loss=0.062836 lr=8.947733e-06 elapsed=09:13:04 avg_step=1.33s recent_step=1.66s eta=27:40:58 +2026-08-25 13:08:45 | INFO | train rank=0 | step=24990/100000 progress=25.0% loss=0.044189 lr=8.946718e-06 elapsed=09:13:15 avg_step=1.33s recent_step=1.15s eta=27:40:39 +2026-08-25 13:08:58 | INFO | train rank=0 | step=25000/100000 progress=25.0% loss=0.068359 lr=8.945703e-06 elapsed=09:13:27 avg_step=1.33s recent_step=1.23s eta=27:40:23 +2026-08-25 13:09:12 | INFO | train rank=0 | step=25010/100000 progress=25.0% loss=0.080444 lr=8.944687e-06 elapsed=09:13:42 avg_step=1.33s recent_step=1.45s eta=27:40:14 +2026-08-25 13:09:25 | INFO | train rank=0 | step=25020/100000 progress=25.0% loss=0.066345 lr=8.943671e-06 elapsed=09:13:55 avg_step=1.33s recent_step=1.26s eta=27:39:58 +2026-08-25 13:09:41 | INFO | train rank=0 | step=25030/100000 progress=25.0% loss=0.055603 lr=8.942654e-06 elapsed=09:14:11 avg_step=1.33s recent_step=1.66s eta=27:39:55 +2026-08-25 13:09:52 | INFO | train rank=0 | step=25040/100000 progress=25.0% loss=0.051331 lr=8.941637e-06 elapsed=09:14:22 avg_step=1.33s recent_step=1.12s eta=27:39:35 +2026-08-25 13:10:05 | INFO | train rank=0 | step=25050/100000 progress=25.1% loss=0.082336 lr=8.940619e-06 elapsed=09:14:35 avg_step=1.33s recent_step=1.27s eta=27:39:20 +2026-08-25 13:10:19 | INFO | train rank=0 | step=25060/100000 progress=25.1% loss=0.048309 lr=8.939601e-06 elapsed=09:14:49 avg_step=1.33s recent_step=1.37s eta=27:39:08 +2026-08-25 13:10:32 | INFO | train rank=0 | step=25070/100000 progress=25.1% loss=0.042786 lr=8.938583e-06 elapsed=09:15:02 avg_step=1.33s recent_step=1.34s eta=27:38:55 +2026-08-25 13:10:44 | INFO | train rank=0 | step=25080/100000 progress=25.1% loss=0.088440 lr=8.937564e-06 elapsed=09:15:14 avg_step=1.33s recent_step=1.21s eta=27:38:39 +2026-08-25 13:10:58 | INFO | train rank=0 | step=25090/100000 progress=25.1% loss=0.066101 lr=8.936545e-06 elapsed=09:15:28 avg_step=1.33s recent_step=1.37s eta=27:38:27 +2026-08-25 13:11:10 | INFO | train rank=0 | step=25100/100000 progress=25.1% loss=0.051865 lr=8.935525e-06 elapsed=09:15:39 avg_step=1.33s recent_step=1.14s eta=27:38:08 +2026-08-25 13:11:26 | INFO | train rank=0 | step=25110/100000 progress=25.1% loss=0.063721 lr=8.934505e-06 elapsed=09:15:56 avg_step=1.33s recent_step=1.66s eta=27:38:05 +2026-08-25 13:11:44 | INFO | train rank=0 | step=25120/100000 progress=25.1% loss=0.046021 lr=8.933485e-06 elapsed=09:16:14 avg_step=1.33s recent_step=1.83s eta=27:38:06 +2026-08-25 13:12:01 | INFO | train rank=0 | step=25130/100000 progress=25.1% loss=0.078857 lr=8.932464e-06 elapsed=09:16:31 avg_step=1.33s recent_step=1.62s eta=27:38:02 +2026-08-25 13:12:12 | INFO | train rank=0 | step=25140/100000 progress=25.1% loss=0.076538 lr=8.931442e-06 elapsed=09:16:42 avg_step=1.33s recent_step=1.13s eta=27:37:43 +2026-08-25 13:12:23 | INFO | train rank=0 | step=25150/100000 progress=25.1% loss=0.085510 lr=8.930420e-06 elapsed=09:16:53 avg_step=1.33s recent_step=1.14s eta=27:37:24 +2026-08-25 13:12:38 | INFO | train rank=0 | step=25160/100000 progress=25.2% loss=0.059341 lr=8.929398e-06 elapsed=09:17:08 avg_step=1.33s recent_step=1.50s eta=27:37:16 +2026-08-25 13:12:53 | INFO | train rank=0 | step=25170/100000 progress=25.2% loss=0.037186 lr=8.928375e-06 elapsed=09:17:22 avg_step=1.33s recent_step=1.42s eta=27:37:05 +2026-08-25 13:13:07 | INFO | train rank=0 | step=25180/100000 progress=25.2% loss=0.052589 lr=8.927352e-06 elapsed=09:17:37 avg_step=1.33s recent_step=1.44s eta=27:36:55 +2026-08-25 13:13:19 | INFO | train rank=0 | step=25190/100000 progress=25.2% loss=0.050201 lr=8.926329e-06 elapsed=09:17:49 avg_step=1.33s recent_step=1.19s eta=27:36:37 +2026-08-25 13:13:30 | INFO | train rank=0 | step=25200/100000 progress=25.2% loss=0.061661 lr=8.925305e-06 elapsed=09:18:00 avg_step=1.33s recent_step=1.11s eta=27:36:18 +2026-08-25 13:13:43 | INFO | train rank=0 | step=25210/100000 progress=25.2% loss=0.066467 lr=8.924280e-06 elapsed=09:18:13 avg_step=1.33s recent_step=1.27s eta=27:36:03 +2026-08-25 13:14:01 | INFO | train rank=0 | step=25220/100000 progress=25.2% loss=0.058777 lr=8.923256e-06 elapsed=09:18:31 avg_step=1.33s recent_step=1.81s eta=27:36:04 +2026-08-25 13:14:12 | INFO | train rank=0 | step=25230/100000 progress=25.2% loss=0.036972 lr=8.922230e-06 elapsed=09:18:42 avg_step=1.33s recent_step=1.12s eta=27:35:44 +2026-08-25 13:14:25 | INFO | train rank=0 | step=25240/100000 progress=25.2% loss=0.065643 lr=8.921205e-06 elapsed=09:18:55 avg_step=1.33s recent_step=1.34s eta=27:35:31 +2026-08-25 13:14:39 | INFO | train rank=0 | step=25250/100000 progress=25.2% loss=0.057373 lr=8.920178e-06 elapsed=09:19:09 avg_step=1.33s recent_step=1.40s eta=27:35:20 +2026-08-25 13:14:52 | INFO | train rank=0 | step=25260/100000 progress=25.3% loss=0.042908 lr=8.919152e-06 elapsed=09:19:22 avg_step=1.33s recent_step=1.25s eta=27:35:04 +2026-08-25 13:15:04 | INFO | train rank=0 | step=25270/100000 progress=25.3% loss=0.056580 lr=8.918125e-06 elapsed=09:19:34 avg_step=1.33s recent_step=1.18s eta=27:34:47 +2026-08-25 13:15:15 | INFO | train rank=0 | step=25280/100000 progress=25.3% loss=0.069275 lr=8.917098e-06 elapsed=09:19:45 avg_step=1.33s recent_step=1.17s eta=27:34:29 +2026-08-25 13:15:28 | INFO | train rank=0 | step=25290/100000 progress=25.3% loss=0.044861 lr=8.916070e-06 elapsed=09:19:58 avg_step=1.33s recent_step=1.28s eta=27:34:14 +2026-08-25 13:15:40 | INFO | train rank=0 | step=25300/100000 progress=25.3% loss=0.054459 lr=8.915041e-06 elapsed=09:20:09 avg_step=1.33s recent_step=1.14s eta=27:33:55 +2026-08-25 13:15:54 | INFO | train rank=0 | step=25310/100000 progress=25.3% loss=0.066589 lr=8.914013e-06 elapsed=09:20:24 avg_step=1.33s recent_step=1.49s eta=27:33:47 +2026-08-25 13:16:12 | INFO | train rank=0 | step=25320/100000 progress=25.3% loss=0.051086 lr=8.912984e-06 elapsed=09:20:41 avg_step=1.33s recent_step=1.71s eta=27:33:45 +2026-08-25 13:16:23 | INFO | train rank=0 | step=25330/100000 progress=25.3% loss=0.050186 lr=8.911954e-06 elapsed=09:20:53 avg_step=1.33s recent_step=1.16s eta=27:33:26 +2026-08-25 13:16:34 | INFO | train rank=0 | step=25340/100000 progress=25.3% loss=0.054962 lr=8.910924e-06 elapsed=09:21:04 avg_step=1.33s recent_step=1.12s eta=27:33:07 +2026-08-25 13:16:46 | INFO | train rank=0 | step=25350/100000 progress=25.4% loss=0.049881 lr=8.909894e-06 elapsed=09:21:16 avg_step=1.33s recent_step=1.16s eta=27:32:48 +2026-08-25 13:16:58 | INFO | train rank=0 | step=25360/100000 progress=25.4% loss=0.041763 lr=8.908863e-06 elapsed=09:21:27 avg_step=1.33s recent_step=1.16s eta=27:32:30 +2026-08-25 13:17:13 | INFO | train rank=0 | step=25370/100000 progress=25.4% loss=0.057343 lr=8.907832e-06 elapsed=09:21:43 avg_step=1.33s recent_step=1.52s eta=27:32:23 +2026-08-25 13:17:24 | INFO | train rank=0 | step=25380/100000 progress=25.4% loss=0.065338 lr=8.906800e-06 elapsed=09:21:54 avg_step=1.33s recent_step=1.15s eta=27:32:04 +2026-08-25 13:17:36 | INFO | train rank=0 | step=25390/100000 progress=25.4% loss=0.122009 lr=8.905768e-06 elapsed=09:22:05 avg_step=1.33s recent_step=1.14s eta=27:31:45 +2026-08-25 13:17:50 | INFO | train rank=0 | step=25400/100000 progress=25.4% loss=0.040802 lr=8.904735e-06 elapsed=09:22:20 avg_step=1.33s recent_step=1.45s eta=27:31:35 +2026-08-25 13:18:05 | INFO | train rank=0 | step=25410/100000 progress=25.4% loss=0.048019 lr=8.903702e-06 elapsed=09:22:35 avg_step=1.33s recent_step=1.46s eta=27:31:26 +2026-08-25 13:18:16 | INFO | train rank=0 | step=25420/100000 progress=25.4% loss=0.104828 lr=8.902669e-06 elapsed=09:22:46 avg_step=1.33s recent_step=1.16s eta=27:31:08 +2026-08-25 13:18:32 | INFO | train rank=0 | step=25430/100000 progress=25.4% loss=0.069489 lr=8.901635e-06 elapsed=09:23:02 avg_step=1.33s recent_step=1.60s eta=27:31:03 +2026-08-25 13:18:48 | INFO | train rank=0 | step=25440/100000 progress=25.4% loss=0.045441 lr=8.900601e-06 elapsed=09:23:17 avg_step=1.33s recent_step=1.53s eta=27:30:55 +2026-08-25 13:18:59 | INFO | train rank=0 | step=25450/100000 progress=25.4% loss=0.050980 lr=8.899566e-06 elapsed=09:23:29 avg_step=1.33s recent_step=1.15s eta=27:30:37 +2026-08-25 13:19:12 | INFO | train rank=0 | step=25460/100000 progress=25.5% loss=0.132080 lr=8.898531e-06 elapsed=09:23:42 avg_step=1.33s recent_step=1.29s eta=27:30:22 +2026-08-25 13:19:23 | INFO | train rank=0 | step=25470/100000 progress=25.5% loss=0.037949 lr=8.897496e-06 elapsed=09:23:53 avg_step=1.33s recent_step=1.14s eta=27:30:04 +2026-08-25 13:19:42 | INFO | train rank=0 | step=25480/100000 progress=25.5% loss=0.070923 lr=8.896460e-06 elapsed=09:24:12 avg_step=1.33s recent_step=1.85s eta=27:30:05 +2026-08-25 13:19:53 | INFO | train rank=0 | step=25490/100000 progress=25.5% loss=0.043945 lr=8.895423e-06 elapsed=09:24:23 avg_step=1.33s recent_step=1.13s eta=27:29:46 +2026-08-25 13:20:05 | INFO | train rank=0 | step=25500/100000 progress=25.5% loss=0.047638 lr=8.894386e-06 elapsed=09:24:35 avg_step=1.33s recent_step=1.18s eta=27:29:29 +2026-08-25 13:20:20 | INFO | train rank=0 | step=25510/100000 progress=25.5% loss=0.047195 lr=8.893349e-06 elapsed=09:24:50 avg_step=1.33s recent_step=1.54s eta=27:29:22 +2026-08-25 13:20:39 | INFO | train rank=0 | step=25520/100000 progress=25.5% loss=0.047806 lr=8.892312e-06 elapsed=09:25:09 avg_step=1.33s recent_step=1.88s eta=27:29:25 +2026-08-25 13:20:51 | INFO | train rank=0 | step=25530/100000 progress=25.5% loss=0.046341 lr=8.891273e-06 elapsed=09:25:21 avg_step=1.33s recent_step=1.17s eta=27:29:07 +2026-08-25 13:21:08 | INFO | train rank=0 | step=25540/100000 progress=25.5% loss=0.060852 lr=8.890235e-06 elapsed=09:25:38 avg_step=1.33s recent_step=1.72s eta=27:29:05 +2026-08-25 13:21:20 | INFO | train rank=0 | step=25550/100000 progress=25.6% loss=0.053375 lr=8.889196e-06 elapsed=09:25:50 avg_step=1.33s recent_step=1.17s eta=27:28:47 +2026-08-25 13:21:34 | INFO | train rank=0 | step=25560/100000 progress=25.6% loss=0.058807 lr=8.888157e-06 elapsed=09:26:04 avg_step=1.33s recent_step=1.41s eta=27:28:36 +2026-08-25 13:21:45 | INFO | train rank=0 | step=25570/100000 progress=25.6% loss=0.109802 lr=8.887117e-06 elapsed=09:26:15 avg_step=1.33s recent_step=1.16s eta=27:28:18 +2026-08-25 13:21:57 | INFO | train rank=0 | step=25580/100000 progress=25.6% loss=0.055801 lr=8.886077e-06 elapsed=09:26:27 avg_step=1.33s recent_step=1.18s eta=27:28:00 +2026-08-25 13:22:09 | INFO | train rank=0 | step=25590/100000 progress=25.6% loss=0.096008 lr=8.885036e-06 elapsed=09:26:39 avg_step=1.33s recent_step=1.18s eta=27:27:42 +2026-08-25 13:22:21 | INFO | train rank=0 | step=25600/100000 progress=25.6% loss=0.052917 lr=8.883995e-06 elapsed=09:26:51 avg_step=1.33s recent_step=1.16s eta=27:27:24 +2026-08-25 13:22:33 | INFO | train rank=0 | step=25610/100000 progress=25.6% loss=0.065155 lr=8.882953e-06 elapsed=09:27:03 avg_step=1.33s recent_step=1.24s eta=27:27:08 +2026-08-25 13:22:48 | INFO | train rank=0 | step=25620/100000 progress=25.6% loss=0.061295 lr=8.881912e-06 elapsed=09:27:18 avg_step=1.33s recent_step=1.46s eta=27:26:59 +2026-08-25 13:23:03 | INFO | train rank=0 | step=25630/100000 progress=25.6% loss=0.101196 lr=8.880869e-06 elapsed=09:27:33 avg_step=1.33s recent_step=1.53s eta=27:26:52 +2026-08-25 13:23:16 | INFO | train rank=0 | step=25640/100000 progress=25.6% loss=0.049774 lr=8.879826e-06 elapsed=09:27:46 avg_step=1.33s recent_step=1.33s eta=27:26:38 +2026-08-25 13:23:33 | INFO | train rank=0 | step=25650/100000 progress=25.7% loss=0.084808 lr=8.878783e-06 elapsed=09:28:03 avg_step=1.33s recent_step=1.69s eta=27:26:36 +2026-08-25 13:23:45 | INFO | train rank=0 | step=25660/100000 progress=25.7% loss=0.060730 lr=8.877740e-06 elapsed=09:28:14 avg_step=1.33s recent_step=1.13s eta=27:26:16 +2026-08-25 13:23:59 | INFO | train rank=0 | step=25670/100000 progress=25.7% loss=0.058685 lr=8.876696e-06 elapsed=09:28:29 avg_step=1.33s recent_step=1.43s eta=27:26:06 +2026-08-25 13:24:11 | INFO | train rank=0 | step=25680/100000 progress=25.7% loss=0.052200 lr=8.875651e-06 elapsed=09:28:41 avg_step=1.33s recent_step=1.18s eta=27:25:49 +2026-08-25 13:24:24 | INFO | train rank=0 | step=25690/100000 progress=25.7% loss=0.046494 lr=8.874606e-06 elapsed=09:28:54 avg_step=1.33s recent_step=1.35s eta=27:25:36 +2026-08-25 13:24:40 | INFO | train rank=0 | step=25700/100000 progress=25.7% loss=0.040207 lr=8.873561e-06 elapsed=09:29:09 avg_step=1.33s recent_step=1.54s eta=27:25:29 +2026-08-25 13:24:51 | INFO | train rank=0 | step=25710/100000 progress=25.7% loss=0.048424 lr=8.872515e-06 elapsed=09:29:21 avg_step=1.33s recent_step=1.17s eta=27:25:11 +2026-08-25 13:25:10 | INFO | train rank=0 | step=25720/100000 progress=25.7% loss=0.072586 lr=8.871469e-06 elapsed=09:29:40 avg_step=1.33s recent_step=1.87s eta=27:25:13 +2026-08-25 13:25:22 | INFO | train rank=0 | step=25730/100000 progress=25.7% loss=0.044449 lr=8.870423e-06 elapsed=09:29:52 avg_step=1.33s recent_step=1.19s eta=27:24:56 +2026-08-25 13:25:35 | INFO | train rank=0 | step=25740/100000 progress=25.7% loss=0.044113 lr=8.869376e-06 elapsed=09:30:05 avg_step=1.33s recent_step=1.33s eta=27:24:42 +2026-08-25 13:25:46 | INFO | train rank=0 | step=25750/100000 progress=25.8% loss=0.041916 lr=8.868328e-06 elapsed=09:30:16 avg_step=1.33s recent_step=1.13s eta=27:24:23 +2026-08-25 13:25:58 | INFO | train rank=0 | step=25760/100000 progress=25.8% loss=0.083221 lr=8.867280e-06 elapsed=09:30:28 avg_step=1.33s recent_step=1.20s eta=27:24:06 +2026-08-25 13:26:10 | INFO | train rank=0 | step=25770/100000 progress=25.8% loss=0.055573 lr=8.866232e-06 elapsed=09:30:40 avg_step=1.33s recent_step=1.16s eta=27:23:48 +2026-08-25 13:26:22 | INFO | train rank=0 | step=25780/100000 progress=25.8% loss=0.140442 lr=8.865183e-06 elapsed=09:30:52 avg_step=1.33s recent_step=1.24s eta=27:23:32 +2026-08-25 13:26:37 | INFO | train rank=0 | step=25790/100000 progress=25.8% loss=0.086487 lr=8.864134e-06 elapsed=09:31:07 avg_step=1.33s recent_step=1.47s eta=27:23:23 +2026-08-25 13:26:49 | INFO | train rank=0 | step=25800/100000 progress=25.8% loss=0.048920 lr=8.863085e-06 elapsed=09:31:18 avg_step=1.33s recent_step=1.15s eta=27:23:05 +2026-08-25 13:27:00 | INFO | train rank=0 | step=25810/100000 progress=25.8% loss=0.108612 lr=8.862035e-06 elapsed=09:31:30 avg_step=1.33s recent_step=1.18s eta=27:22:47 +2026-08-25 13:27:14 | INFO | train rank=0 | step=25820/100000 progress=25.8% loss=0.066162 lr=8.860984e-06 elapsed=09:31:44 avg_step=1.33s recent_step=1.38s eta=27:22:35 +2026-08-25 13:27:26 | INFO | train rank=0 | step=25830/100000 progress=25.8% loss=0.073608 lr=8.859934e-06 elapsed=09:31:56 avg_step=1.33s recent_step=1.21s eta=27:22:18 +2026-08-25 13:27:43 | INFO | train rank=0 | step=25840/100000 progress=25.8% loss=0.042664 lr=8.858882e-06 elapsed=09:32:13 avg_step=1.33s recent_step=1.67s eta=27:22:15 +2026-08-25 13:27:57 | INFO | train rank=0 | step=25850/100000 progress=25.9% loss=0.045822 lr=8.857831e-06 elapsed=09:32:27 avg_step=1.33s recent_step=1.43s eta=27:22:05 +2026-08-25 13:28:11 | INFO | train rank=0 | step=25860/100000 progress=25.9% loss=0.087585 lr=8.856779e-06 elapsed=09:32:41 avg_step=1.33s recent_step=1.41s eta=27:21:54 +2026-08-25 13:28:23 | INFO | train rank=0 | step=25870/100000 progress=25.9% loss=0.075104 lr=8.855726e-06 elapsed=09:32:53 avg_step=1.33s recent_step=1.15s eta=27:21:35 +2026-08-25 13:28:34 | INFO | train rank=0 | step=25880/100000 progress=25.9% loss=0.075592 lr=8.854673e-06 elapsed=09:33:04 avg_step=1.33s recent_step=1.16s eta=27:21:17 +2026-08-25 13:28:51 | INFO | train rank=0 | step=25890/100000 progress=25.9% loss=0.060120 lr=8.853620e-06 elapsed=09:33:21 avg_step=1.33s recent_step=1.65s eta=27:21:13 +2026-08-25 13:29:02 | INFO | train rank=0 | step=25900/100000 progress=25.9% loss=0.058411 lr=8.852566e-06 elapsed=09:33:32 avg_step=1.33s recent_step=1.13s eta=27:20:54 +2026-08-25 13:29:14 | INFO | train rank=0 | step=25910/100000 progress=25.9% loss=0.048965 lr=8.851512e-06 elapsed=09:33:44 avg_step=1.33s recent_step=1.19s eta=27:20:36 +2026-08-25 13:29:26 | INFO | train rank=0 | step=25920/100000 progress=25.9% loss=0.081360 lr=8.850457e-06 elapsed=09:33:56 avg_step=1.33s recent_step=1.18s eta=27:20:19 +2026-08-25 13:29:37 | INFO | train rank=0 | step=25930/100000 progress=25.9% loss=0.039536 lr=8.849402e-06 elapsed=09:34:07 avg_step=1.33s recent_step=1.17s eta=27:20:01 +2026-08-25 13:29:55 | INFO | train rank=0 | step=25940/100000 progress=25.9% loss=0.062500 lr=8.848347e-06 elapsed=09:34:25 avg_step=1.33s recent_step=1.80s eta=27:20:01 +2026-08-25 13:30:07 | INFO | train rank=0 | step=25950/100000 progress=25.9% loss=0.049591 lr=8.847291e-06 elapsed=09:34:37 avg_step=1.33s recent_step=1.15s eta=27:19:43 +2026-08-25 13:30:21 | INFO | train rank=0 | step=25960/100000 progress=26.0% loss=0.044250 lr=8.846235e-06 elapsed=09:34:51 avg_step=1.33s recent_step=1.37s eta=27:19:31 +2026-08-25 13:30:33 | INFO | train rank=0 | step=25970/100000 progress=26.0% loss=0.031631 lr=8.845178e-06 elapsed=09:35:03 avg_step=1.33s recent_step=1.19s eta=27:19:14 +2026-08-25 13:30:44 | INFO | train rank=0 | step=25980/100000 progress=26.0% loss=0.038986 lr=8.844121e-06 elapsed=09:35:14 avg_step=1.33s recent_step=1.15s eta=27:18:55 +2026-08-25 13:30:58 | INFO | train rank=0 | step=25990/100000 progress=26.0% loss=0.071350 lr=8.843064e-06 elapsed=09:35:28 avg_step=1.33s recent_step=1.38s eta=27:18:44 +2026-08-25 13:31:13 | INFO | train rank=0 | step=26000/100000 progress=26.0% loss=0.056244 lr=8.842006e-06 elapsed=09:35:42 avg_step=1.33s recent_step=1.45s eta=27:18:34 +2026-08-25 13:31:26 | INFO | train rank=0 | step=26010/100000 progress=26.0% loss=0.079834 lr=8.840947e-06 elapsed=09:35:56 avg_step=1.33s recent_step=1.38s eta=27:18:22 +2026-08-25 13:31:41 | INFO | train rank=0 | step=26020/100000 progress=26.0% loss=0.046082 lr=8.839888e-06 elapsed=09:36:11 avg_step=1.33s recent_step=1.45s eta=27:18:12 +2026-08-25 13:31:52 | INFO | train rank=0 | step=26030/100000 progress=26.0% loss=0.053253 lr=8.838829e-06 elapsed=09:36:22 avg_step=1.33s recent_step=1.14s eta=27:17:54 +2026-08-25 13:32:04 | INFO | train rank=0 | step=26040/100000 progress=26.0% loss=0.058899 lr=8.837770e-06 elapsed=09:36:34 avg_step=1.33s recent_step=1.18s eta=27:17:36 +2026-08-25 13:32:16 | INFO | train rank=0 | step=26050/100000 progress=26.1% loss=0.053543 lr=8.836709e-06 elapsed=09:36:46 avg_step=1.33s recent_step=1.23s eta=27:17:20 +2026-08-25 13:32:30 | INFO | train rank=0 | step=26060/100000 progress=26.1% loss=0.044708 lr=8.835649e-06 elapsed=09:37:00 avg_step=1.33s recent_step=1.34s eta=27:17:07 +2026-08-25 13:32:42 | INFO | train rank=0 | step=26070/100000 progress=26.1% loss=0.047089 lr=8.834588e-06 elapsed=09:37:12 avg_step=1.33s recent_step=1.24s eta=27:16:51 +2026-08-25 13:32:55 | INFO | train rank=0 | step=26080/100000 progress=26.1% loss=0.034515 lr=8.833527e-06 elapsed=09:37:25 avg_step=1.33s recent_step=1.31s eta=27:16:38 +2026-08-25 13:33:11 | INFO | train rank=0 | step=26090/100000 progress=26.1% loss=0.054031 lr=8.832465e-06 elapsed=09:37:41 avg_step=1.33s recent_step=1.58s eta=27:16:31 +2026-08-25 13:33:24 | INFO | train rank=0 | step=26100/100000 progress=26.1% loss=0.062653 lr=8.831403e-06 elapsed=09:37:53 avg_step=1.33s recent_step=1.25s eta=27:16:16 +2026-08-25 13:33:42 | INFO | train rank=0 | step=26110/100000 progress=26.1% loss=0.093567 lr=8.830340e-06 elapsed=09:38:12 avg_step=1.33s recent_step=1.82s eta=27:16:16 +2026-08-25 13:33:54 | INFO | train rank=0 | step=26120/100000 progress=26.1% loss=0.094238 lr=8.829277e-06 elapsed=09:38:24 avg_step=1.33s recent_step=1.21s eta=27:16:00 +2026-08-25 13:34:06 | INFO | train rank=0 | step=26130/100000 progress=26.1% loss=0.071442 lr=8.828214e-06 elapsed=09:38:36 avg_step=1.33s recent_step=1.18s eta=27:15:42 +2026-08-25 13:34:19 | INFO | train rank=0 | step=26140/100000 progress=26.1% loss=0.061127 lr=8.827150e-06 elapsed=09:38:49 avg_step=1.33s recent_step=1.32s eta=27:15:29 +2026-08-25 13:34:32 | INFO | train rank=0 | step=26150/100000 progress=26.2% loss=0.065460 lr=8.826086e-06 elapsed=09:39:02 avg_step=1.33s recent_step=1.34s eta=27:15:16 +2026-08-25 13:34:44 | INFO | train rank=0 | step=26160/100000 progress=26.2% loss=0.064606 lr=8.825021e-06 elapsed=09:39:14 avg_step=1.33s recent_step=1.18s eta=27:14:59 +2026-08-25 13:34:56 | INFO | train rank=0 | step=26170/100000 progress=26.2% loss=0.036057 lr=8.823956e-06 elapsed=09:39:26 avg_step=1.33s recent_step=1.18s eta=27:14:41 +2026-08-25 13:35:08 | INFO | train rank=0 | step=26180/100000 progress=26.2% loss=0.066681 lr=8.822891e-06 elapsed=09:39:38 avg_step=1.33s recent_step=1.25s eta=27:14:26 +2026-08-25 13:35:23 | INFO | train rank=0 | step=26190/100000 progress=26.2% loss=0.056961 lr=8.821825e-06 elapsed=09:39:53 avg_step=1.33s recent_step=1.49s eta=27:14:17 +2026-08-25 13:35:35 | INFO | train rank=0 | step=26200/100000 progress=26.2% loss=0.121704 lr=8.820758e-06 elapsed=09:40:05 avg_step=1.33s recent_step=1.17s eta=27:13:59 +2026-08-25 13:35:49 | INFO | train rank=0 | step=26210/100000 progress=26.2% loss=0.062927 lr=8.819692e-06 elapsed=09:40:19 avg_step=1.33s recent_step=1.42s eta=27:13:49 +2026-08-25 13:36:04 | INFO | train rank=0 | step=26220/100000 progress=26.2% loss=0.052078 lr=8.818624e-06 elapsed=09:40:34 avg_step=1.33s recent_step=1.48s eta=27:13:40 +2026-08-25 13:36:16 | INFO | train rank=0 | step=26230/100000 progress=26.2% loss=0.045380 lr=8.817557e-06 elapsed=09:40:46 avg_step=1.33s recent_step=1.18s eta=27:13:22 +2026-08-25 13:36:28 | INFO | train rank=0 | step=26240/100000 progress=26.2% loss=0.044907 lr=8.816489e-06 elapsed=09:40:58 avg_step=1.33s recent_step=1.21s eta=27:13:06 +2026-08-25 13:36:42 | INFO | train rank=0 | step=26250/100000 progress=26.2% loss=0.052444 lr=8.815420e-06 elapsed=09:41:12 avg_step=1.33s recent_step=1.40s eta=27:12:54 +2026-08-25 13:36:57 | INFO | train rank=0 | step=26260/100000 progress=26.3% loss=0.088745 lr=8.814351e-06 elapsed=09:41:27 avg_step=1.33s recent_step=1.53s eta=27:12:47 +2026-08-25 13:37:12 | INFO | train rank=0 | step=26270/100000 progress=26.3% loss=0.049500 lr=8.813282e-06 elapsed=09:41:42 avg_step=1.33s recent_step=1.50s eta=27:12:38 +2026-08-25 13:37:24 | INFO | train rank=0 | step=26280/100000 progress=26.3% loss=0.067902 lr=8.812213e-06 elapsed=09:41:54 avg_step=1.33s recent_step=1.19s eta=27:12:21 +2026-08-25 13:37:38 | INFO | train rank=0 | step=26290/100000 progress=26.3% loss=0.061417 lr=8.811142e-06 elapsed=09:42:08 avg_step=1.33s recent_step=1.35s eta=27:12:09 +2026-08-25 13:37:51 | INFO | train rank=0 | step=26300/100000 progress=26.3% loss=0.053864 lr=8.810072e-06 elapsed=09:42:21 avg_step=1.33s recent_step=1.37s eta=27:11:57 +2026-08-25 13:38:03 | INFO | train rank=0 | step=26310/100000 progress=26.3% loss=0.053329 lr=8.809001e-06 elapsed=09:42:33 avg_step=1.33s recent_step=1.18s eta=27:11:39 +2026-08-25 13:38:17 | INFO | train rank=0 | step=26320/100000 progress=26.3% loss=0.059494 lr=8.807930e-06 elapsed=09:42:47 avg_step=1.33s recent_step=1.36s eta=27:11:27 +2026-08-25 13:38:29 | INFO | train rank=0 | step=26330/100000 progress=26.3% loss=0.159729 lr=8.806858e-06 elapsed=09:42:59 avg_step=1.33s recent_step=1.20s eta=27:11:10 +2026-08-25 13:38:45 | INFO | train rank=0 | step=26340/100000 progress=26.3% loss=0.078735 lr=8.805786e-06 elapsed=09:43:15 avg_step=1.33s recent_step=1.63s eta=27:11:05 +2026-08-25 13:38:59 | INFO | train rank=0 | step=26350/100000 progress=26.4% loss=0.061249 lr=8.804713e-06 elapsed=09:43:29 avg_step=1.33s recent_step=1.40s eta=27:10:54 +2026-08-25 13:39:11 | INFO | train rank=0 | step=26360/100000 progress=26.4% loss=0.045502 lr=8.803640e-06 elapsed=09:43:41 avg_step=1.33s recent_step=1.19s eta=27:10:36 +2026-08-25 13:39:23 | INFO | train rank=0 | step=26370/100000 progress=26.4% loss=0.056564 lr=8.802567e-06 elapsed=09:43:52 avg_step=1.33s recent_step=1.15s eta=27:10:18 +2026-08-25 13:39:40 | INFO | train rank=0 | step=26380/100000 progress=26.4% loss=0.049255 lr=8.801493e-06 elapsed=09:44:09 avg_step=1.33s recent_step=1.69s eta=27:10:15 +2026-08-25 13:39:57 | INFO | train rank=0 | step=26390/100000 progress=26.4% loss=0.064636 lr=8.800419e-06 elapsed=09:44:27 avg_step=1.33s recent_step=1.74s eta=27:10:13 +2026-08-25 13:40:08 | INFO | train rank=0 | step=26400/100000 progress=26.4% loss=0.061829 lr=8.799344e-06 elapsed=09:44:38 avg_step=1.33s recent_step=1.13s eta=27:09:54 +2026-08-25 13:40:21 | INFO | train rank=0 | step=26410/100000 progress=26.4% loss=0.042572 lr=8.798269e-06 elapsed=09:44:51 avg_step=1.33s recent_step=1.28s eta=27:09:40 +2026-08-25 13:40:36 | INFO | train rank=0 | step=26420/100000 progress=26.4% loss=0.071655 lr=8.797193e-06 elapsed=09:45:06 avg_step=1.33s recent_step=1.53s eta=27:09:32 +2026-08-25 13:40:47 | INFO | train rank=0 | step=26430/100000 progress=26.4% loss=0.065002 lr=8.796117e-06 elapsed=09:45:17 avg_step=1.33s recent_step=1.12s eta=27:09:13 +2026-08-25 13:40:59 | INFO | train rank=0 | step=26440/100000 progress=26.4% loss=0.074432 lr=8.795041e-06 elapsed=09:45:29 avg_step=1.33s recent_step=1.17s eta=27:08:55 +2026-08-25 13:41:13 | INFO | train rank=0 | step=26450/100000 progress=26.5% loss=0.040680 lr=8.793964e-06 elapsed=09:45:43 avg_step=1.33s recent_step=1.38s eta=27:08:43 +2026-08-25 13:41:25 | INFO | train rank=0 | step=26460/100000 progress=26.5% loss=0.044914 lr=8.792887e-06 elapsed=09:45:55 avg_step=1.33s recent_step=1.18s eta=27:08:26 +2026-08-25 13:41:37 | INFO | train rank=0 | step=26470/100000 progress=26.5% loss=0.060150 lr=8.791810e-06 elapsed=09:46:07 avg_step=1.33s recent_step=1.19s eta=27:08:09 +2026-08-25 13:41:50 | INFO | train rank=0 | step=26480/100000 progress=26.5% loss=0.039963 lr=8.790732e-06 elapsed=09:46:20 avg_step=1.33s recent_step=1.30s eta=27:07:55 +2026-08-25 13:42:04 | INFO | train rank=0 | step=26490/100000 progress=26.5% loss=0.062683 lr=8.789653e-06 elapsed=09:46:34 avg_step=1.33s recent_step=1.42s eta=27:07:44 +2026-08-25 13:42:17 | INFO | train rank=0 | step=26500/100000 progress=26.5% loss=0.078888 lr=8.788574e-06 elapsed=09:46:46 avg_step=1.33s recent_step=1.27s eta=27:07:29 +2026-08-25 13:42:28 | INFO | train rank=0 | step=26510/100000 progress=26.5% loss=0.066437 lr=8.787495e-06 elapsed=09:46:58 avg_step=1.33s recent_step=1.14s eta=27:07:11 +2026-08-25 13:42:43 | INFO | train rank=0 | step=26520/100000 progress=26.5% loss=0.056839 lr=8.786415e-06 elapsed=09:47:13 avg_step=1.33s recent_step=1.54s eta=27:07:03 +2026-08-25 13:42:55 | INFO | train rank=0 | step=26530/100000 progress=26.5% loss=0.073029 lr=8.785335e-06 elapsed=09:47:25 avg_step=1.33s recent_step=1.15s eta=27:06:45 +2026-08-25 13:43:09 | INFO | train rank=0 | step=26540/100000 progress=26.5% loss=0.075165 lr=8.784255e-06 elapsed=09:47:39 avg_step=1.33s recent_step=1.39s eta=27:06:34 +2026-08-25 13:43:25 | INFO | train rank=0 | step=26550/100000 progress=26.6% loss=0.047638 lr=8.783174e-06 elapsed=09:47:55 avg_step=1.33s recent_step=1.59s eta=27:06:28 +2026-08-25 13:43:36 | INFO | train rank=0 | step=26560/100000 progress=26.6% loss=0.049179 lr=8.782093e-06 elapsed=09:48:06 avg_step=1.33s recent_step=1.13s eta=27:06:09 +2026-08-25 13:43:48 | INFO | train rank=0 | step=26570/100000 progress=26.6% loss=0.045761 lr=8.781011e-06 elapsed=09:48:18 avg_step=1.33s recent_step=1.17s eta=27:05:51 +2026-08-25 13:43:59 | INFO | train rank=0 | step=26580/100000 progress=26.6% loss=0.051010 lr=8.779929e-06 elapsed=09:48:29 avg_step=1.33s recent_step=1.16s eta=27:05:33 +2026-08-25 13:44:11 | INFO | train rank=0 | step=26590/100000 progress=26.6% loss=0.049927 lr=8.778846e-06 elapsed=09:48:41 avg_step=1.33s recent_step=1.16s eta=27:05:15 +2026-08-25 13:44:26 | INFO | train rank=0 | step=26600/100000 progress=26.6% loss=0.062042 lr=8.777763e-06 elapsed=09:48:56 avg_step=1.33s recent_step=1.52s eta=27:05:07 +2026-08-25 13:44:38 | INFO | train rank=0 | step=26610/100000 progress=26.6% loss=0.073822 lr=8.776680e-06 elapsed=09:49:08 avg_step=1.33s recent_step=1.16s eta=27:04:49 +2026-08-25 13:44:53 | INFO | train rank=0 | step=26620/100000 progress=26.6% loss=0.055008 lr=8.775596e-06 elapsed=09:49:23 avg_step=1.33s recent_step=1.51s eta=27:04:41 +2026-08-25 13:45:06 | INFO | train rank=0 | step=26630/100000 progress=26.6% loss=0.077759 lr=8.774512e-06 elapsed=09:49:36 avg_step=1.33s recent_step=1.28s eta=27:04:26 +2026-08-25 13:45:20 | INFO | train rank=0 | step=26640/100000 progress=26.6% loss=0.056137 lr=8.773427e-06 elapsed=09:49:50 avg_step=1.33s recent_step=1.40s eta=27:04:15 +2026-08-25 13:45:35 | INFO | train rank=0 | step=26650/100000 progress=26.7% loss=0.067841 lr=8.772342e-06 elapsed=09:50:05 avg_step=1.33s recent_step=1.52s eta=27:04:07 +2026-08-25 13:45:49 | INFO | train rank=0 | step=26660/100000 progress=26.7% loss=0.079224 lr=8.771257e-06 elapsed=09:50:19 avg_step=1.33s recent_step=1.39s eta=27:03:56 +2026-08-25 13:46:04 | INFO | train rank=0 | step=26670/100000 progress=26.7% loss=0.051804 lr=8.770171e-06 elapsed=09:50:34 avg_step=1.33s recent_step=1.51s eta=27:03:47 +2026-08-25 13:46:16 | INFO | train rank=0 | step=26680/100000 progress=26.7% loss=0.052628 lr=8.769085e-06 elapsed=09:50:46 avg_step=1.33s recent_step=1.17s eta=27:03:30 +2026-08-25 13:46:29 | INFO | train rank=0 | step=26690/100000 progress=26.7% loss=0.057190 lr=8.767998e-06 elapsed=09:50:59 avg_step=1.33s recent_step=1.35s eta=27:03:17 +2026-08-25 13:46:41 | INFO | train rank=0 | step=26700/100000 progress=26.7% loss=0.064056 lr=8.766911e-06 elapsed=09:51:11 avg_step=1.33s recent_step=1.18s eta=27:03:00 +2026-08-25 13:46:53 | INFO | train rank=0 | step=26710/100000 progress=26.7% loss=0.034172 lr=8.765824e-06 elapsed=09:51:23 avg_step=1.33s recent_step=1.20s eta=27:02:43 +2026-08-25 13:47:05 | INFO | train rank=0 | step=26720/100000 progress=26.7% loss=0.072906 lr=8.764736e-06 elapsed=09:51:35 avg_step=1.33s recent_step=1.17s eta=27:02:25 +2026-08-25 13:47:17 | INFO | train rank=0 | step=26730/100000 progress=26.7% loss=0.067123 lr=8.763647e-06 elapsed=09:51:47 avg_step=1.33s recent_step=1.21s eta=27:02:09 +2026-08-25 13:47:29 | INFO | train rank=0 | step=26740/100000 progress=26.7% loss=0.043488 lr=8.762559e-06 elapsed=09:51:59 avg_step=1.33s recent_step=1.23s eta=27:01:53 +2026-08-25 13:47:42 | INFO | train rank=0 | step=26750/100000 progress=26.8% loss=0.056961 lr=8.761469e-06 elapsed=09:52:12 avg_step=1.33s recent_step=1.33s eta=27:01:40 +2026-08-25 13:47:54 | INFO | train rank=0 | step=26760/100000 progress=26.8% loss=0.061813 lr=8.760380e-06 elapsed=09:52:24 avg_step=1.33s recent_step=1.19s eta=27:01:22 +2026-08-25 13:48:08 | INFO | train rank=0 | step=26770/100000 progress=26.8% loss=0.054077 lr=8.759290e-06 elapsed=09:52:38 avg_step=1.33s recent_step=1.36s eta=27:01:10 +2026-08-25 13:48:20 | INFO | train rank=0 | step=26780/100000 progress=26.8% loss=0.096985 lr=8.758200e-06 elapsed=09:52:50 avg_step=1.33s recent_step=1.19s eta=27:00:53 +2026-08-25 13:48:39 | INFO | train rank=0 | step=26790/100000 progress=26.8% loss=0.143250 lr=8.757109e-06 elapsed=09:53:08 avg_step=1.33s recent_step=1.88s eta=27:00:55 +2026-08-25 13:48:53 | INFO | train rank=0 | step=26800/100000 progress=26.8% loss=0.097198 lr=8.756018e-06 elapsed=09:53:23 avg_step=1.33s recent_step=1.46s eta=27:00:45 +2026-08-25 13:49:07 | INFO | train rank=0 | step=26810/100000 progress=26.8% loss=0.108215 lr=8.754926e-06 elapsed=09:53:37 avg_step=1.33s recent_step=1.43s eta=27:00:34 +2026-08-25 13:49:19 | INFO | train rank=0 | step=26820/100000 progress=26.8% loss=0.028938 lr=8.753834e-06 elapsed=09:53:49 avg_step=1.33s recent_step=1.14s eta=27:00:16 +2026-08-25 13:49:32 | INFO | train rank=0 | step=26830/100000 progress=26.8% loss=0.051483 lr=8.752741e-06 elapsed=09:54:02 avg_step=1.33s recent_step=1.35s eta=27:00:03 +2026-08-25 13:49:44 | INFO | train rank=0 | step=26840/100000 progress=26.8% loss=0.042126 lr=8.751649e-06 elapsed=09:54:14 avg_step=1.33s recent_step=1.21s eta=26:59:47 +2026-08-25 13:49:56 | INFO | train rank=0 | step=26850/100000 progress=26.9% loss=0.054489 lr=8.750555e-06 elapsed=09:54:26 avg_step=1.33s recent_step=1.15s eta=26:59:29 +2026-08-25 13:50:08 | INFO | train rank=0 | step=26860/100000 progress=26.9% loss=0.060974 lr=8.749462e-06 elapsed=09:54:38 avg_step=1.33s recent_step=1.21s eta=26:59:12 +2026-08-25 13:50:23 | INFO | train rank=0 | step=26870/100000 progress=26.9% loss=0.055038 lr=8.748368e-06 elapsed=09:54:53 avg_step=1.33s recent_step=1.51s eta=26:59:04 +2026-08-25 13:50:35 | INFO | train rank=0 | step=26880/100000 progress=26.9% loss=0.040070 lr=8.747273e-06 elapsed=09:55:05 avg_step=1.33s recent_step=1.18s eta=26:58:46 +2026-08-25 13:50:49 | INFO | train rank=0 | step=26890/100000 progress=26.9% loss=0.034409 lr=8.746178e-06 elapsed=09:55:19 avg_step=1.33s recent_step=1.41s eta=26:58:35 +2026-08-25 13:51:01 | INFO | train rank=0 | step=26900/100000 progress=26.9% loss=0.055054 lr=8.745083e-06 elapsed=09:55:31 avg_step=1.33s recent_step=1.21s eta=26:58:19 +2026-08-25 13:51:14 | INFO | train rank=0 | step=26910/100000 progress=26.9% loss=0.040085 lr=8.743987e-06 elapsed=09:55:44 avg_step=1.33s recent_step=1.34s eta=26:58:06 +2026-08-25 13:51:29 | INFO | train rank=0 | step=26920/100000 progress=26.9% loss=0.090454 lr=8.742891e-06 elapsed=09:55:59 avg_step=1.33s recent_step=1.45s eta=26:57:56 +2026-08-25 13:51:44 | INFO | train rank=0 | step=26930/100000 progress=26.9% loss=0.042419 lr=8.741795e-06 elapsed=09:56:14 avg_step=1.33s recent_step=1.53s eta=26:57:48 +2026-08-25 13:51:56 | INFO | train rank=0 | step=26940/100000 progress=26.9% loss=0.049500 lr=8.740698e-06 elapsed=09:56:26 avg_step=1.33s recent_step=1.21s eta=26:57:31 +2026-08-25 13:52:11 | INFO | train rank=0 | step=26950/100000 progress=27.0% loss=0.061859 lr=8.739600e-06 elapsed=09:56:41 avg_step=1.33s recent_step=1.47s eta=26:57:22 +2026-08-25 13:52:23 | INFO | train rank=0 | step=26960/100000 progress=27.0% loss=0.060242 lr=8.738503e-06 elapsed=09:56:53 avg_step=1.33s recent_step=1.17s eta=26:57:05 +2026-08-25 13:52:35 | INFO | train rank=0 | step=26970/100000 progress=27.0% loss=0.053864 lr=8.737404e-06 elapsed=09:57:05 avg_step=1.33s recent_step=1.19s eta=26:56:47 +2026-08-25 13:52:49 | INFO | train rank=0 | step=26980/100000 progress=27.0% loss=0.070648 lr=8.736306e-06 elapsed=09:57:19 avg_step=1.33s recent_step=1.43s eta=26:56:37 +2026-08-25 13:53:05 | INFO | train rank=0 | step=26990/100000 progress=27.0% loss=0.037575 lr=8.735207e-06 elapsed=09:57:35 avg_step=1.33s recent_step=1.58s eta=26:56:30 +2026-08-25 13:53:20 | INFO | train rank=0 | step=27000/100000 progress=27.0% loss=0.067505 lr=8.734107e-06 elapsed=09:57:50 avg_step=1.33s recent_step=1.51s eta=26:56:22 +2026-08-25 13:53:32 | INFO | train rank=0 | step=27010/100000 progress=27.0% loss=0.084839 lr=8.733008e-06 elapsed=09:58:02 avg_step=1.33s recent_step=1.21s eta=26:56:06 +2026-08-25 13:53:44 | INFO | train rank=0 | step=27020/100000 progress=27.0% loss=0.056793 lr=8.731907e-06 elapsed=09:58:14 avg_step=1.33s recent_step=1.18s eta=26:55:49 +2026-08-25 13:53:58 | INFO | train rank=0 | step=27030/100000 progress=27.0% loss=0.044342 lr=8.730807e-06 elapsed=09:58:28 avg_step=1.33s recent_step=1.45s eta=26:55:39 +2026-08-25 13:54:10 | INFO | train rank=0 | step=27040/100000 progress=27.0% loss=0.050629 lr=8.729706e-06 elapsed=09:58:40 avg_step=1.33s recent_step=1.17s eta=26:55:21 +2026-08-25 13:54:22 | INFO | train rank=0 | step=27050/100000 progress=27.1% loss=0.044510 lr=8.728604e-06 elapsed=09:58:52 avg_step=1.33s recent_step=1.19s eta=26:55:04 +2026-08-25 13:54:33 | INFO | train rank=0 | step=27060/100000 progress=27.1% loss=0.056595 lr=8.727502e-06 elapsed=09:59:03 avg_step=1.33s recent_step=1.14s eta=26:54:46 +2026-08-25 13:54:48 | INFO | train rank=0 | step=27070/100000 progress=27.1% loss=0.047058 lr=8.726400e-06 elapsed=09:59:18 avg_step=1.33s recent_step=1.47s eta=26:54:36 +2026-08-25 13:55:00 | INFO | train rank=0 | step=27080/100000 progress=27.1% loss=0.101501 lr=8.725298e-06 elapsed=09:59:29 avg_step=1.33s recent_step=1.15s eta=26:54:18 +2026-08-25 13:55:13 | INFO | train rank=0 | step=27090/100000 progress=27.1% loss=0.054840 lr=8.724194e-06 elapsed=09:59:42 avg_step=1.33s recent_step=1.30s eta=26:54:04 +2026-08-25 13:55:29 | INFO | train rank=0 | step=27100/100000 progress=27.1% loss=0.083069 lr=8.723091e-06 elapsed=09:59:59 avg_step=1.33s recent_step=1.67s eta=26:54:00 +2026-08-25 13:55:41 | INFO | train rank=0 | step=27110/100000 progress=27.1% loss=0.052734 lr=8.721987e-06 elapsed=10:00:11 avg_step=1.33s recent_step=1.16s eta=26:53:42 +2026-08-25 13:55:55 | INFO | train rank=0 | step=27120/100000 progress=27.1% loss=0.069977 lr=8.720883e-06 elapsed=10:00:25 avg_step=1.33s recent_step=1.44s eta=26:53:32 +2026-08-25 13:56:07 | INFO | train rank=0 | step=27130/100000 progress=27.1% loss=0.063522 lr=8.719778e-06 elapsed=10:00:37 avg_step=1.33s recent_step=1.20s eta=26:53:15 +2026-08-25 13:56:22 | INFO | train rank=0 | step=27140/100000 progress=27.1% loss=0.055542 lr=8.718673e-06 elapsed=10:00:51 avg_step=1.33s recent_step=1.43s eta=26:53:05 +2026-08-25 13:56:37 | INFO | train rank=0 | step=27150/100000 progress=27.2% loss=0.038712 lr=8.717568e-06 elapsed=10:01:07 avg_step=1.33s recent_step=1.54s eta=26:52:57 +2026-08-25 13:56:49 | INFO | train rank=0 | step=27160/100000 progress=27.2% loss=0.070526 lr=8.716462e-06 elapsed=10:01:19 avg_step=1.33s recent_step=1.17s eta=26:52:39 +2026-08-25 13:57:04 | INFO | train rank=0 | step=27170/100000 progress=27.2% loss=0.051590 lr=8.715355e-06 elapsed=10:01:34 avg_step=1.33s recent_step=1.50s eta=26:52:31 +2026-08-25 13:57:18 | INFO | train rank=0 | step=27180/100000 progress=27.2% loss=0.054108 lr=8.714249e-06 elapsed=10:01:48 avg_step=1.33s recent_step=1.40s eta=26:52:19 +2026-08-25 13:57:30 | INFO | train rank=0 | step=27190/100000 progress=27.2% loss=0.043900 lr=8.713141e-06 elapsed=10:01:59 avg_step=1.33s recent_step=1.18s eta=26:52:02 +2026-08-25 13:57:41 | INFO | train rank=0 | step=27200/100000 progress=27.2% loss=0.050400 lr=8.712034e-06 elapsed=10:02:11 avg_step=1.33s recent_step=1.17s eta=26:51:45 +2026-08-25 13:57:53 | INFO | train rank=0 | step=27210/100000 progress=27.2% loss=0.051392 lr=8.710926e-06 elapsed=10:02:22 avg_step=1.33s recent_step=1.13s eta=26:51:26 +2026-08-25 13:58:05 | INFO | train rank=0 | step=27220/100000 progress=27.2% loss=0.073303 lr=8.709818e-06 elapsed=10:02:34 avg_step=1.33s recent_step=1.20s eta=26:51:09 +2026-08-25 13:58:21 | INFO | train rank=0 | step=27230/100000 progress=27.2% loss=0.073395 lr=8.708709e-06 elapsed=10:02:51 avg_step=1.33s recent_step=1.63s eta=26:51:04 +2026-08-25 13:58:33 | INFO | train rank=0 | step=27240/100000 progress=27.2% loss=0.062286 lr=8.707600e-06 elapsed=10:03:03 avg_step=1.33s recent_step=1.20s eta=26:50:48 +2026-08-25 13:58:48 | INFO | train rank=0 | step=27250/100000 progress=27.3% loss=0.083923 lr=8.706490e-06 elapsed=10:03:18 avg_step=1.33s recent_step=1.49s eta=26:50:39 +2026-08-25 13:58:59 | INFO | train rank=0 | step=27260/100000 progress=27.3% loss=0.057678 lr=8.705380e-06 elapsed=10:03:29 avg_step=1.33s recent_step=1.15s eta=26:50:21 +2026-08-25 13:59:14 | INFO | train rank=0 | step=27270/100000 progress=27.3% loss=0.052521 lr=8.704270e-06 elapsed=10:03:44 avg_step=1.33s recent_step=1.48s eta=26:50:11 +2026-08-25 13:59:29 | INFO | train rank=0 | step=27280/100000 progress=27.3% loss=0.053345 lr=8.703159e-06 elapsed=10:03:59 avg_step=1.33s recent_step=1.47s eta=26:50:02 +2026-08-25 13:59:41 | INFO | train rank=0 | step=27290/100000 progress=27.3% loss=0.058777 lr=8.702048e-06 elapsed=10:04:10 avg_step=1.33s recent_step=1.18s eta=26:49:45 +2026-08-25 13:59:53 | INFO | train rank=0 | step=27300/100000 progress=27.3% loss=0.074860 lr=8.700936e-06 elapsed=10:04:23 avg_step=1.33s recent_step=1.25s eta=26:49:29 +2026-08-25 14:00:08 | INFO | train rank=0 | step=27310/100000 progress=27.3% loss=0.105835 lr=8.699824e-06 elapsed=10:04:38 avg_step=1.33s recent_step=1.51s eta=26:49:21 +2026-08-25 14:00:21 | INFO | train rank=0 | step=27320/100000 progress=27.3% loss=0.055145 lr=8.698712e-06 elapsed=10:04:50 avg_step=1.33s recent_step=1.23s eta=26:49:05 +2026-08-25 14:00:35 | INFO | train rank=0 | step=27330/100000 progress=27.3% loss=0.119080 lr=8.697599e-06 elapsed=10:05:05 avg_step=1.33s recent_step=1.49s eta=26:48:56 +2026-08-25 14:00:47 | INFO | train rank=0 | step=27340/100000 progress=27.3% loss=0.043701 lr=8.696486e-06 elapsed=10:05:16 avg_step=1.33s recent_step=1.12s eta=26:48:37 +2026-08-25 14:00:58 | INFO | train rank=0 | step=27350/100000 progress=27.4% loss=0.071381 lr=8.695372e-06 elapsed=10:05:28 avg_step=1.33s recent_step=1.18s eta=26:48:20 +2026-08-25 14:01:14 | INFO | train rank=0 | step=27360/100000 progress=27.4% loss=0.077393 lr=8.694258e-06 elapsed=10:05:44 avg_step=1.33s recent_step=1.54s eta=26:48:12 +2026-08-25 14:01:26 | INFO | train rank=0 | step=27370/100000 progress=27.4% loss=0.059784 lr=8.693144e-06 elapsed=10:05:56 avg_step=1.33s recent_step=1.20s eta=26:47:55 +2026-08-25 14:01:37 | INFO | train rank=0 | step=27380/100000 progress=27.4% loss=0.099548 lr=8.692029e-06 elapsed=10:06:07 avg_step=1.33s recent_step=1.15s eta=26:47:37 +2026-08-25 14:01:49 | INFO | train rank=0 | step=27390/100000 progress=27.4% loss=0.051208 lr=8.690914e-06 elapsed=10:06:19 avg_step=1.33s recent_step=1.15s eta=26:47:19 +2026-08-25 14:02:01 | INFO | train rank=0 | step=27400/100000 progress=27.4% loss=0.072052 lr=8.689798e-06 elapsed=10:06:31 avg_step=1.33s recent_step=1.24s eta=26:47:03 +2026-08-25 14:02:14 | INFO | train rank=0 | step=27410/100000 progress=27.4% loss=0.063507 lr=8.688682e-06 elapsed=10:06:44 avg_step=1.33s recent_step=1.31s eta=26:46:50 +2026-08-25 14:02:29 | INFO | train rank=0 | step=27420/100000 progress=27.4% loss=0.049583 lr=8.687566e-06 elapsed=10:06:59 avg_step=1.33s recent_step=1.46s eta=26:46:40 +2026-08-25 14:02:45 | INFO | train rank=0 | step=27430/100000 progress=27.4% loss=0.074066 lr=8.686449e-06 elapsed=10:07:15 avg_step=1.33s recent_step=1.66s eta=26:46:35 +2026-08-25 14:02:57 | INFO | train rank=0 | step=27440/100000 progress=27.4% loss=0.072693 lr=8.685331e-06 elapsed=10:07:27 avg_step=1.33s recent_step=1.17s eta=26:46:18 +2026-08-25 14:03:09 | INFO | train rank=0 | step=27450/100000 progress=27.5% loss=0.107483 lr=8.684214e-06 elapsed=10:07:39 avg_step=1.33s recent_step=1.22s eta=26:46:02 +2026-08-25 14:03:21 | INFO | train rank=0 | step=27460/100000 progress=27.5% loss=0.048920 lr=8.683096e-06 elapsed=10:07:51 avg_step=1.33s recent_step=1.16s eta=26:45:44 +2026-08-25 14:03:36 | INFO | train rank=0 | step=27470/100000 progress=27.5% loss=0.058411 lr=8.681977e-06 elapsed=10:08:06 avg_step=1.33s recent_step=1.52s eta=26:45:36 +2026-08-25 14:03:52 | INFO | train rank=0 | step=27480/100000 progress=27.5% loss=0.043716 lr=8.680858e-06 elapsed=10:08:22 avg_step=1.33s recent_step=1.63s eta=26:45:30 +2026-08-25 14:04:04 | INFO | train rank=0 | step=27490/100000 progress=27.5% loss=0.058502 lr=8.679739e-06 elapsed=10:08:33 avg_step=1.33s recent_step=1.13s eta=26:45:12 +2026-08-25 14:04:18 | INFO | train rank=0 | step=27500/100000 progress=27.5% loss=0.078369 lr=8.678620e-06 elapsed=10:08:48 avg_step=1.33s recent_step=1.43s eta=26:45:01 +2026-08-25 14:04:33 | INFO | train rank=0 | step=27510/100000 progress=27.5% loss=0.057526 lr=8.677499e-06 elapsed=10:09:02 avg_step=1.33s recent_step=1.48s eta=26:44:52 +2026-08-25 14:04:47 | INFO | train rank=0 | step=27520/100000 progress=27.5% loss=0.076889 lr=8.676379e-06 elapsed=10:09:17 avg_step=1.33s recent_step=1.43s eta=26:44:41 +2026-08-25 14:04:58 | INFO | train rank=0 | step=27530/100000 progress=27.5% loss=0.082077 lr=8.675258e-06 elapsed=10:09:28 avg_step=1.33s recent_step=1.15s eta=26:44:23 +2026-08-25 14:05:13 | INFO | train rank=0 | step=27540/100000 progress=27.5% loss=0.042206 lr=8.674137e-06 elapsed=10:09:43 avg_step=1.33s recent_step=1.50s eta=26:44:14 +2026-08-25 14:05:25 | INFO | train rank=0 | step=27550/100000 progress=27.6% loss=0.047485 lr=8.673015e-06 elapsed=10:09:55 avg_step=1.33s recent_step=1.15s eta=26:43:56 +2026-08-25 14:05:41 | INFO | train rank=0 | step=27560/100000 progress=27.6% loss=0.046478 lr=8.671893e-06 elapsed=10:10:11 avg_step=1.33s recent_step=1.62s eta=26:43:51 +2026-08-25 14:05:52 | INFO | train rank=0 | step=27570/100000 progress=27.6% loss=0.058472 lr=8.670771e-06 elapsed=10:10:22 avg_step=1.33s recent_step=1.14s eta=26:43:32 +2026-08-25 14:06:06 | INFO | train rank=0 | step=27580/100000 progress=27.6% loss=0.079407 lr=8.669648e-06 elapsed=10:10:36 avg_step=1.33s recent_step=1.40s eta=26:43:21 +2026-08-25 14:06:21 | INFO | train rank=0 | step=27590/100000 progress=27.6% loss=0.060806 lr=8.668524e-06 elapsed=10:10:51 avg_step=1.33s recent_step=1.47s eta=26:43:12 +2026-08-25 14:06:37 | INFO | train rank=0 | step=27600/100000 progress=27.6% loss=0.067673 lr=8.667401e-06 elapsed=10:11:07 avg_step=1.33s recent_step=1.58s eta=26:43:05 +2026-08-25 14:06:49 | INFO | train rank=0 | step=27610/100000 progress=27.6% loss=0.046890 lr=8.666277e-06 elapsed=10:11:19 avg_step=1.33s recent_step=1.17s eta=26:42:47 +2026-08-25 14:07:02 | INFO | train rank=0 | step=27620/100000 progress=27.6% loss=0.083740 lr=8.665152e-06 elapsed=10:11:32 avg_step=1.33s recent_step=1.32s eta=26:42:34 +2026-08-25 14:07:18 | INFO | train rank=0 | step=27630/100000 progress=27.6% loss=0.047127 lr=8.664027e-06 elapsed=10:11:48 avg_step=1.33s recent_step=1.61s eta=26:42:28 +2026-08-25 14:07:30 | INFO | train rank=0 | step=27640/100000 progress=27.6% loss=0.054382 lr=8.662902e-06 elapsed=10:12:00 avg_step=1.33s recent_step=1.19s eta=26:42:11 +2026-08-25 14:07:41 | INFO | train rank=0 | step=27650/100000 progress=27.7% loss=0.034111 lr=8.661776e-06 elapsed=10:12:11 avg_step=1.33s recent_step=1.13s eta=26:41:53 +2026-08-25 14:07:53 | INFO | train rank=0 | step=27660/100000 progress=27.7% loss=0.069733 lr=8.660650e-06 elapsed=10:12:23 avg_step=1.33s recent_step=1.17s eta=26:41:36 +2026-08-25 14:08:04 | INFO | train rank=0 | step=27670/100000 progress=27.7% loss=0.040543 lr=8.659524e-06 elapsed=10:12:34 avg_step=1.33s recent_step=1.14s eta=26:41:17 +2026-08-25 14:08:16 | INFO | train rank=0 | step=27680/100000 progress=27.7% loss=0.080200 lr=8.658397e-06 elapsed=10:12:46 avg_step=1.33s recent_step=1.13s eta=26:40:59 +2026-08-25 14:08:30 | INFO | train rank=0 | step=27690/100000 progress=27.7% loss=0.049744 lr=8.657270e-06 elapsed=10:13:00 avg_step=1.33s recent_step=1.45s eta=26:40:49 +2026-08-25 14:08:48 | INFO | train rank=0 | step=27700/100000 progress=27.7% loss=0.056732 lr=8.656142e-06 elapsed=10:13:18 avg_step=1.33s recent_step=1.77s eta=26:40:47 +2026-08-25 14:09:02 | INFO | train rank=0 | step=27710/100000 progress=27.7% loss=0.048553 lr=8.655014e-06 elapsed=10:13:32 avg_step=1.33s recent_step=1.45s eta=26:40:37 +2026-08-25 14:09:16 | INFO | train rank=0 | step=27720/100000 progress=27.7% loss=0.070129 lr=8.653885e-06 elapsed=10:13:46 avg_step=1.33s recent_step=1.38s eta=26:40:25 +2026-08-25 14:09:27 | INFO | train rank=0 | step=27730/100000 progress=27.7% loss=0.075394 lr=8.652757e-06 elapsed=10:13:57 avg_step=1.33s recent_step=1.10s eta=26:40:06 +2026-08-25 14:09:39 | INFO | train rank=0 | step=27740/100000 progress=27.7% loss=0.051804 lr=8.651627e-06 elapsed=10:14:09 avg_step=1.33s recent_step=1.19s eta=26:39:49 +2026-08-25 14:09:52 | INFO | train rank=0 | step=27750/100000 progress=27.8% loss=0.062988 lr=8.650498e-06 elapsed=10:14:22 avg_step=1.33s recent_step=1.29s eta=26:39:34 +2026-08-25 14:10:06 | INFO | train rank=0 | step=27760/100000 progress=27.8% loss=0.048096 lr=8.649367e-06 elapsed=10:14:36 avg_step=1.33s recent_step=1.39s eta=26:39:23 +2026-08-25 14:10:21 | INFO | train rank=0 | step=27770/100000 progress=27.8% loss=0.090576 lr=8.648237e-06 elapsed=10:14:50 avg_step=1.33s recent_step=1.47s eta=26:39:13 +2026-08-25 14:10:34 | INFO | train rank=0 | step=27780/100000 progress=27.8% loss=0.044647 lr=8.647106e-06 elapsed=10:15:04 avg_step=1.33s recent_step=1.33s eta=26:39:00 +2026-08-25 14:10:46 | INFO | train rank=0 | step=27790/100000 progress=27.8% loss=0.073761 lr=8.645975e-06 elapsed=10:15:16 avg_step=1.33s recent_step=1.17s eta=26:38:43 +2026-08-25 14:10:59 | INFO | train rank=0 | step=27800/100000 progress=27.8% loss=0.097168 lr=8.644843e-06 elapsed=10:15:29 avg_step=1.33s recent_step=1.31s eta=26:38:29 +2026-08-25 14:11:13 | INFO | train rank=0 | step=27810/100000 progress=27.8% loss=0.064102 lr=8.643711e-06 elapsed=10:15:43 avg_step=1.33s recent_step=1.42s eta=26:38:18 +2026-08-25 14:11:29 | INFO | train rank=0 | step=27820/100000 progress=27.8% loss=0.060898 lr=8.642579e-06 elapsed=10:15:59 avg_step=1.33s recent_step=1.58s eta=26:38:11 +2026-08-25 14:11:40 | INFO | train rank=0 | step=27830/100000 progress=27.8% loss=0.047974 lr=8.641446e-06 elapsed=10:16:10 avg_step=1.33s recent_step=1.13s eta=26:37:53 +2026-08-25 14:11:56 | INFO | train rank=0 | step=27840/100000 progress=27.8% loss=0.057999 lr=8.640312e-06 elapsed=10:16:25 avg_step=1.33s recent_step=1.55s eta=26:37:45 +2026-08-25 14:12:07 | INFO | train rank=0 | step=27850/100000 progress=27.9% loss=0.069092 lr=8.639179e-06 elapsed=10:16:37 avg_step=1.33s recent_step=1.18s eta=26:37:28 +2026-08-25 14:12:20 | INFO | train rank=0 | step=27860/100000 progress=27.9% loss=0.094421 lr=8.638045e-06 elapsed=10:16:50 avg_step=1.33s recent_step=1.29s eta=26:37:14 +2026-08-25 14:12:34 | INFO | train rank=0 | step=27870/100000 progress=27.9% loss=0.088013 lr=8.636910e-06 elapsed=10:17:04 avg_step=1.33s recent_step=1.39s eta=26:37:02 +2026-08-25 14:12:48 | INFO | train rank=0 | step=27880/100000 progress=27.9% loss=0.076202 lr=8.635775e-06 elapsed=10:17:18 avg_step=1.33s recent_step=1.40s eta=26:36:51 +2026-08-25 14:13:01 | INFO | train rank=0 | step=27890/100000 progress=27.9% loss=0.045883 lr=8.634640e-06 elapsed=10:17:31 avg_step=1.33s recent_step=1.33s eta=26:36:37 +2026-08-25 14:13:15 | INFO | train rank=0 | step=27900/100000 progress=27.9% loss=0.049637 lr=8.633504e-06 elapsed=10:17:45 avg_step=1.33s recent_step=1.41s eta=26:36:26 +2026-08-25 14:13:29 | INFO | train rank=0 | step=27910/100000 progress=27.9% loss=0.047394 lr=8.632368e-06 elapsed=10:17:58 avg_step=1.33s recent_step=1.32s eta=26:36:12 +2026-08-25 14:13:44 | INFO | train rank=0 | step=27920/100000 progress=27.9% loss=0.058960 lr=8.631232e-06 elapsed=10:18:14 avg_step=1.33s recent_step=1.51s eta=26:36:04 +2026-08-25 14:13:56 | INFO | train rank=0 | step=27930/100000 progress=27.9% loss=0.051270 lr=8.630095e-06 elapsed=10:18:25 avg_step=1.33s recent_step=1.19s eta=26:35:47 +2026-08-25 14:14:07 | INFO | train rank=0 | step=27940/100000 progress=27.9% loss=0.042343 lr=8.628958e-06 elapsed=10:18:37 avg_step=1.33s recent_step=1.14s eta=26:35:29 +2026-08-25 14:14:21 | INFO | train rank=0 | step=27950/100000 progress=28.0% loss=0.095337 lr=8.627820e-06 elapsed=10:18:51 avg_step=1.33s recent_step=1.44s eta=26:35:18 +2026-08-25 14:14:37 | INFO | train rank=0 | step=27960/100000 progress=28.0% loss=0.081482 lr=8.626682e-06 elapsed=10:19:07 avg_step=1.33s recent_step=1.56s eta=26:35:11 +2026-08-25 14:14:51 | INFO | train rank=0 | step=27970/100000 progress=28.0% loss=0.052902 lr=8.625544e-06 elapsed=10:19:21 avg_step=1.33s recent_step=1.43s eta=26:35:00 +2026-08-25 14:15:04 | INFO | train rank=0 | step=27980/100000 progress=28.0% loss=0.057419 lr=8.624405e-06 elapsed=10:19:34 avg_step=1.33s recent_step=1.24s eta=26:34:45 +2026-08-25 14:15:18 | INFO | train rank=0 | step=27990/100000 progress=28.0% loss=0.061920 lr=8.623266e-06 elapsed=10:19:48 avg_step=1.33s recent_step=1.44s eta=26:34:34 +2026-08-25 14:15:30 | INFO | train rank=0 | step=28000/100000 progress=28.0% loss=0.047287 lr=8.622126e-06 elapsed=10:20:00 avg_step=1.33s recent_step=1.17s eta=26:34:17 +2026-08-25 14:15:42 | INFO | train rank=0 | step=28010/100000 progress=28.0% loss=0.031288 lr=8.620986e-06 elapsed=10:20:11 avg_step=1.33s recent_step=1.18s eta=26:34:00 +2026-08-25 14:15:53 | INFO | train rank=0 | step=28020/100000 progress=28.0% loss=0.049164 lr=8.619846e-06 elapsed=10:20:23 avg_step=1.33s recent_step=1.15s eta=26:33:42 +2026-08-25 14:16:07 | INFO | train rank=0 | step=28030/100000 progress=28.0% loss=0.059418 lr=8.618705e-06 elapsed=10:20:37 avg_step=1.33s recent_step=1.36s eta=26:33:30 +2026-08-25 14:16:19 | INFO | train rank=0 | step=28040/100000 progress=28.0% loss=0.058075 lr=8.617564e-06 elapsed=10:20:49 avg_step=1.33s recent_step=1.22s eta=26:33:14 +2026-08-25 14:16:31 | INFO | train rank=0 | step=28050/100000 progress=28.1% loss=0.039711 lr=8.616422e-06 elapsed=10:21:01 avg_step=1.33s recent_step=1.22s eta=26:32:58 +2026-08-25 14:16:46 | INFO | train rank=0 | step=28060/100000 progress=28.1% loss=0.068207 lr=8.615280e-06 elapsed=10:21:16 avg_step=1.33s recent_step=1.51s eta=26:32:49 +2026-08-25 14:17:01 | INFO | train rank=0 | step=28070/100000 progress=28.1% loss=0.043556 lr=8.614138e-06 elapsed=10:21:31 avg_step=1.33s recent_step=1.46s eta=26:32:39 +2026-08-25 14:17:12 | INFO | train rank=0 | step=28080/100000 progress=28.1% loss=0.074677 lr=8.612995e-06 elapsed=10:21:42 avg_step=1.33s recent_step=1.16s eta=26:32:22 +2026-08-25 14:17:25 | INFO | train rank=0 | step=28090/100000 progress=28.1% loss=0.059219 lr=8.611852e-06 elapsed=10:21:55 avg_step=1.33s recent_step=1.23s eta=26:32:06 +2026-08-25 14:17:40 | INFO | train rank=0 | step=28100/100000 progress=28.1% loss=0.044037 lr=8.610708e-06 elapsed=10:22:10 avg_step=1.33s recent_step=1.58s eta=26:31:59 +2026-08-25 14:17:54 | INFO | train rank=0 | step=28110/100000 progress=28.1% loss=0.061676 lr=8.609564e-06 elapsed=10:22:24 avg_step=1.33s recent_step=1.38s eta=26:31:47 +2026-08-25 14:18:09 | INFO | train rank=0 | step=28120/100000 progress=28.1% loss=0.035950 lr=8.608420e-06 elapsed=10:22:38 avg_step=1.33s recent_step=1.43s eta=26:31:36 +2026-08-25 14:18:23 | INFO | train rank=0 | step=28130/100000 progress=28.1% loss=0.039734 lr=8.607275e-06 elapsed=10:22:53 avg_step=1.33s recent_step=1.44s eta=26:31:25 +2026-08-25 14:18:37 | INFO | train rank=0 | step=28140/100000 progress=28.1% loss=0.044098 lr=8.606130e-06 elapsed=10:23:07 avg_step=1.33s recent_step=1.40s eta=26:31:14 +2026-08-25 14:18:54 | INFO | train rank=0 | step=28150/100000 progress=28.1% loss=0.157898 lr=8.604984e-06 elapsed=10:23:23 avg_step=1.33s recent_step=1.66s eta=26:31:09 +2026-08-25 14:19:06 | INFO | train rank=0 | step=28160/100000 progress=28.2% loss=0.054474 lr=8.603838e-06 elapsed=10:23:36 avg_step=1.33s recent_step=1.28s eta=26:30:55 +2026-08-25 14:19:18 | INFO | train rank=0 | step=28170/100000 progress=28.2% loss=0.032516 lr=8.602692e-06 elapsed=10:23:48 avg_step=1.33s recent_step=1.15s eta=26:30:37 +2026-08-25 14:19:29 | INFO | train rank=0 | step=28180/100000 progress=28.2% loss=0.043777 lr=8.601545e-06 elapsed=10:23:59 avg_step=1.33s recent_step=1.12s eta=26:30:18 +2026-08-25 14:19:43 | INFO | train rank=0 | step=28190/100000 progress=28.2% loss=0.038391 lr=8.600398e-06 elapsed=10:24:13 avg_step=1.33s recent_step=1.42s eta=26:30:08 +2026-08-25 14:19:56 | INFO | train rank=0 | step=28200/100000 progress=28.2% loss=0.067184 lr=8.599250e-06 elapsed=10:24:26 avg_step=1.33s recent_step=1.30s eta=26:29:54 +2026-08-25 14:20:08 | INFO | train rank=0 | step=28210/100000 progress=28.2% loss=0.046204 lr=8.598103e-06 elapsed=10:24:38 avg_step=1.33s recent_step=1.17s eta=26:29:36 +2026-08-25 14:20:20 | INFO | train rank=0 | step=28220/100000 progress=28.2% loss=0.063538 lr=8.596954e-06 elapsed=10:24:50 avg_step=1.33s recent_step=1.19s eta=26:29:19 +2026-08-25 14:20:34 | INFO | train rank=0 | step=28230/100000 progress=28.2% loss=0.057648 lr=8.595806e-06 elapsed=10:25:04 avg_step=1.33s recent_step=1.37s eta=26:29:07 +2026-08-25 14:20:45 | INFO | train rank=0 | step=28240/100000 progress=28.2% loss=0.098206 lr=8.594656e-06 elapsed=10:25:15 avg_step=1.33s recent_step=1.16s eta=26:28:50 +2026-08-25 14:20:57 | INFO | train rank=0 | step=28250/100000 progress=28.2% loss=0.048676 lr=8.593507e-06 elapsed=10:25:26 avg_step=1.33s recent_step=1.13s eta=26:28:31 +2026-08-25 14:21:12 | INFO | train rank=0 | step=28260/100000 progress=28.3% loss=0.033279 lr=8.592357e-06 elapsed=10:25:42 avg_step=1.33s recent_step=1.54s eta=26:28:23 +2026-08-25 14:21:28 | INFO | train rank=0 | step=28270/100000 progress=28.3% loss=0.039825 lr=8.591207e-06 elapsed=10:25:58 avg_step=1.33s recent_step=1.59s eta=26:28:17 +2026-08-25 14:21:40 | INFO | train rank=0 | step=28280/100000 progress=28.3% loss=0.045517 lr=8.590056e-06 elapsed=10:26:10 avg_step=1.33s recent_step=1.22s eta=26:28:01 +2026-08-25 14:21:52 | INFO | train rank=0 | step=28290/100000 progress=28.3% loss=0.048752 lr=8.588905e-06 elapsed=10:26:22 avg_step=1.33s recent_step=1.19s eta=26:27:44 +2026-08-25 14:22:04 | INFO | train rank=0 | step=28300/100000 progress=28.3% loss=0.057083 lr=8.587754e-06 elapsed=10:26:34 avg_step=1.33s recent_step=1.19s eta=26:27:27 +2026-08-25 14:22:18 | INFO | train rank=0 | step=28310/100000 progress=28.3% loss=0.060455 lr=8.586602e-06 elapsed=10:26:48 avg_step=1.33s recent_step=1.42s eta=26:27:16 +2026-08-25 14:22:30 | INFO | train rank=0 | step=28320/100000 progress=28.3% loss=0.053284 lr=8.585449e-06 elapsed=10:27:00 avg_step=1.33s recent_step=1.19s eta=26:26:59 +2026-08-25 14:22:42 | INFO | train rank=0 | step=28330/100000 progress=28.3% loss=0.064880 lr=8.584297e-06 elapsed=10:27:11 avg_step=1.33s recent_step=1.15s eta=26:26:42 +2026-08-25 14:22:59 | INFO | train rank=0 | step=28340/100000 progress=28.3% loss=0.068298 lr=8.583144e-06 elapsed=10:27:29 avg_step=1.33s recent_step=1.79s eta=26:26:40 +2026-08-25 14:23:11 | INFO | train rank=0 | step=28350/100000 progress=28.3% loss=0.039688 lr=8.581990e-06 elapsed=10:27:41 avg_step=1.33s recent_step=1.16s eta=26:26:22 +2026-08-25 14:23:25 | INFO | train rank=0 | step=28360/100000 progress=28.4% loss=0.062286 lr=8.580837e-06 elapsed=10:27:55 avg_step=1.33s recent_step=1.44s eta=26:26:12 +2026-08-25 14:23:38 | INFO | train rank=0 | step=28370/100000 progress=28.4% loss=0.036072 lr=8.579682e-06 elapsed=10:28:07 avg_step=1.33s recent_step=1.22s eta=26:25:56 +2026-08-25 14:23:52 | INFO | train rank=0 | step=28380/100000 progress=28.4% loss=0.109314 lr=8.578528e-06 elapsed=10:28:22 avg_step=1.33s recent_step=1.47s eta=26:25:46 +2026-08-25 14:24:07 | INFO | train rank=0 | step=28390/100000 progress=28.4% loss=0.082642 lr=8.577373e-06 elapsed=10:28:37 avg_step=1.33s recent_step=1.49s eta=26:25:37 +2026-08-25 14:24:21 | INFO | train rank=0 | step=28400/100000 progress=28.4% loss=0.052582 lr=8.576217e-06 elapsed=10:28:51 avg_step=1.33s recent_step=1.42s eta=26:25:26 +2026-08-25 14:24:35 | INFO | train rank=0 | step=28410/100000 progress=28.4% loss=0.047333 lr=8.575062e-06 elapsed=10:29:05 avg_step=1.33s recent_step=1.33s eta=26:25:13 +2026-08-25 14:24:50 | INFO | train rank=0 | step=28420/100000 progress=28.4% loss=0.070343 lr=8.573906e-06 elapsed=10:29:19 avg_step=1.33s recent_step=1.50s eta=26:25:04 +2026-08-25 14:25:03 | INFO | train rank=0 | step=28430/100000 progress=28.4% loss=0.056000 lr=8.572749e-06 elapsed=10:29:33 avg_step=1.33s recent_step=1.35s eta=26:24:51 +2026-08-25 14:25:16 | INFO | train rank=0 | step=28440/100000 progress=28.4% loss=0.051941 lr=8.571592e-06 elapsed=10:29:45 avg_step=1.33s recent_step=1.24s eta=26:24:36 +2026-08-25 14:25:28 | INFO | train rank=0 | step=28450/100000 progress=28.4% loss=0.038940 lr=8.570435e-06 elapsed=10:29:58 avg_step=1.33s recent_step=1.24s eta=26:24:20 +2026-08-25 14:25:43 | INFO | train rank=0 | step=28460/100000 progress=28.5% loss=0.077927 lr=8.569277e-06 elapsed=10:30:13 avg_step=1.33s recent_step=1.47s eta=26:24:10 +2026-08-25 14:25:58 | INFO | train rank=0 | step=28470/100000 progress=28.5% loss=0.073608 lr=8.568119e-06 elapsed=10:30:27 avg_step=1.33s recent_step=1.49s eta=26:24:01 +2026-08-25 14:26:14 | INFO | train rank=0 | step=28480/100000 progress=28.5% loss=0.058716 lr=8.566960e-06 elapsed=10:30:44 avg_step=1.33s recent_step=1.68s eta=26:23:57 +2026-08-25 14:26:25 | INFO | train rank=0 | step=28490/100000 progress=28.5% loss=0.072021 lr=8.565802e-06 elapsed=10:30:55 avg_step=1.33s recent_step=1.11s eta=26:23:38 +2026-08-25 14:26:37 | INFO | train rank=0 | step=28500/100000 progress=28.5% loss=0.089417 lr=8.564642e-06 elapsed=10:31:07 avg_step=1.33s recent_step=1.19s eta=26:23:21 +2026-08-25 14:26:49 | INFO | train rank=0 | step=28510/100000 progress=28.5% loss=0.052277 lr=8.563483e-06 elapsed=10:31:19 avg_step=1.33s recent_step=1.17s eta=26:23:04 +2026-08-25 14:27:02 | INFO | train rank=0 | step=28520/100000 progress=28.5% loss=0.058777 lr=8.562323e-06 elapsed=10:31:32 avg_step=1.33s recent_step=1.28s eta=26:22:49 +2026-08-25 14:27:14 | INFO | train rank=0 | step=28530/100000 progress=28.5% loss=0.031601 lr=8.561162e-06 elapsed=10:31:44 avg_step=1.33s recent_step=1.18s eta=26:22:32 +2026-08-25 14:27:25 | INFO | train rank=0 | step=28540/100000 progress=28.5% loss=0.039383 lr=8.560001e-06 elapsed=10:31:55 avg_step=1.33s recent_step=1.13s eta=26:22:14 +2026-08-25 14:27:37 | INFO | train rank=0 | step=28550/100000 progress=28.5% loss=0.041977 lr=8.558840e-06 elapsed=10:32:07 avg_step=1.33s recent_step=1.21s eta=26:21:58 +2026-08-25 14:27:55 | INFO | train rank=0 | step=28560/100000 progress=28.6% loss=0.051743 lr=8.557678e-06 elapsed=10:32:25 avg_step=1.33s recent_step=1.79s eta=26:21:56 +2026-08-25 14:28:07 | INFO | train rank=0 | step=28570/100000 progress=28.6% loss=0.086792 lr=8.556516e-06 elapsed=10:32:36 avg_step=1.33s recent_step=1.16s eta=26:21:39 +2026-08-25 14:28:18 | INFO | train rank=0 | step=28580/100000 progress=28.6% loss=0.128662 lr=8.555354e-06 elapsed=10:32:48 avg_step=1.33s recent_step=1.16s eta=26:21:21 +2026-08-25 14:28:29 | INFO | train rank=0 | step=28590/100000 progress=28.6% loss=0.073090 lr=8.554191e-06 elapsed=10:32:59 avg_step=1.33s recent_step=1.12s eta=26:21:03 +2026-08-25 14:28:41 | INFO | train rank=0 | step=28600/100000 progress=28.6% loss=0.078491 lr=8.553028e-06 elapsed=10:33:11 avg_step=1.33s recent_step=1.16s eta=26:20:45 +2026-08-25 14:28:54 | INFO | train rank=0 | step=28610/100000 progress=28.6% loss=0.040558 lr=8.551864e-06 elapsed=10:33:24 avg_step=1.33s recent_step=1.34s eta=26:20:32 +2026-08-25 14:29:10 | INFO | train rank=0 | step=28620/100000 progress=28.6% loss=0.050171 lr=8.550701e-06 elapsed=10:33:40 avg_step=1.33s recent_step=1.52s eta=26:20:24 +2026-08-25 14:29:22 | INFO | train rank=0 | step=28630/100000 progress=28.6% loss=0.056015 lr=8.549536e-06 elapsed=10:33:51 avg_step=1.33s recent_step=1.19s eta=26:20:07 +2026-08-25 14:29:36 | INFO | train rank=0 | step=28640/100000 progress=28.6% loss=0.049072 lr=8.548371e-06 elapsed=10:34:06 avg_step=1.33s recent_step=1.41s eta=26:19:56 +2026-08-25 14:29:48 | INFO | train rank=0 | step=28650/100000 progress=28.6% loss=0.041275 lr=8.547206e-06 elapsed=10:34:18 avg_step=1.33s recent_step=1.22s eta=26:19:40 +2026-08-25 14:30:04 | INFO | train rank=0 | step=28660/100000 progress=28.7% loss=0.069092 lr=8.546041e-06 elapsed=10:34:33 avg_step=1.33s recent_step=1.56s eta=26:19:32 +2026-08-25 14:30:15 | INFO | train rank=0 | step=28670/100000 progress=28.7% loss=0.085510 lr=8.544875e-06 elapsed=10:34:45 avg_step=1.33s recent_step=1.17s eta=26:19:15 +2026-08-25 14:30:27 | INFO | train rank=0 | step=28680/100000 progress=28.7% loss=0.050232 lr=8.543709e-06 elapsed=10:34:57 avg_step=1.33s recent_step=1.17s eta=26:18:58 +2026-08-25 14:30:41 | INFO | train rank=0 | step=28690/100000 progress=28.7% loss=0.068878 lr=8.542542e-06 elapsed=10:35:11 avg_step=1.33s recent_step=1.42s eta=26:18:47 +2026-08-25 14:30:58 | INFO | train rank=0 | step=28700/100000 progress=28.7% loss=0.055695 lr=8.541375e-06 elapsed=10:35:28 avg_step=1.33s recent_step=1.68s eta=26:18:43 +2026-08-25 14:31:13 | INFO | train rank=0 | step=28710/100000 progress=28.7% loss=0.048981 lr=8.540208e-06 elapsed=10:35:43 avg_step=1.33s recent_step=1.51s eta=26:18:34 +2026-08-25 14:31:29 | INFO | train rank=0 | step=28720/100000 progress=28.7% loss=0.091949 lr=8.539040e-06 elapsed=10:35:59 avg_step=1.33s recent_step=1.62s eta=26:18:28 +2026-08-25 14:31:41 | INFO | train rank=0 | step=28730/100000 progress=28.7% loss=0.071533 lr=8.537871e-06 elapsed=10:36:11 avg_step=1.33s recent_step=1.15s eta=26:18:10 +2026-08-25 14:31:53 | INFO | train rank=0 | step=28740/100000 progress=28.7% loss=0.044212 lr=8.536703e-06 elapsed=10:36:23 avg_step=1.33s recent_step=1.23s eta=26:17:54 +2026-08-25 14:32:06 | INFO | train rank=0 | step=28750/100000 progress=28.7% loss=0.047760 lr=8.535534e-06 elapsed=10:36:36 avg_step=1.33s recent_step=1.34s eta=26:17:41 +2026-08-25 14:32:23 | INFO | train rank=0 | step=28760/100000 progress=28.8% loss=0.072845 lr=8.534365e-06 elapsed=10:36:52 avg_step=1.33s recent_step=1.61s eta=26:17:35 +2026-08-25 14:32:34 | INFO | train rank=0 | step=28770/100000 progress=28.8% loss=0.043884 lr=8.533195e-06 elapsed=10:37:04 avg_step=1.33s recent_step=1.16s eta=26:17:18 +2026-08-25 14:32:47 | INFO | train rank=0 | step=28780/100000 progress=28.8% loss=0.044128 lr=8.532025e-06 elapsed=10:37:17 avg_step=1.33s recent_step=1.26s eta=26:17:03 +2026-08-25 14:32:59 | INFO | train rank=0 | step=28790/100000 progress=28.8% loss=0.071320 lr=8.530854e-06 elapsed=10:37:29 avg_step=1.33s recent_step=1.21s eta=26:16:47 +2026-08-25 14:33:11 | INFO | train rank=0 | step=28800/100000 progress=28.8% loss=0.029602 lr=8.529683e-06 elapsed=10:37:41 avg_step=1.33s recent_step=1.20s eta=26:16:30 +2026-08-25 14:33:27 | INFO | train rank=0 | step=28810/100000 progress=28.8% loss=0.073563 lr=8.528512e-06 elapsed=10:37:57 avg_step=1.33s recent_step=1.58s eta=26:16:23 +2026-08-25 14:33:39 | INFO | train rank=0 | step=28820/100000 progress=28.8% loss=0.056396 lr=8.527340e-06 elapsed=10:38:09 avg_step=1.33s recent_step=1.21s eta=26:16:07 +2026-08-25 14:33:53 | INFO | train rank=0 | step=28830/100000 progress=28.8% loss=0.067810 lr=8.526168e-06 elapsed=10:38:23 avg_step=1.33s recent_step=1.38s eta=26:15:55 +2026-08-25 14:34:04 | INFO | train rank=0 | step=28840/100000 progress=28.8% loss=0.066849 lr=8.524996e-06 elapsed=10:38:34 avg_step=1.33s recent_step=1.18s eta=26:15:38 +2026-08-25 14:34:21 | INFO | train rank=0 | step=28850/100000 progress=28.8% loss=0.052643 lr=8.523823e-06 elapsed=10:38:51 avg_step=1.33s recent_step=1.66s eta=26:15:33 +2026-08-25 14:34:33 | INFO | train rank=0 | step=28860/100000 progress=28.9% loss=0.054337 lr=8.522650e-06 elapsed=10:39:02 avg_step=1.33s recent_step=1.15s eta=26:15:15 +2026-08-25 14:34:46 | INFO | train rank=0 | step=28870/100000 progress=28.9% loss=0.046753 lr=8.521476e-06 elapsed=10:39:16 avg_step=1.33s recent_step=1.39s eta=26:15:03 +2026-08-25 14:35:01 | INFO | train rank=0 | step=28880/100000 progress=28.9% loss=0.039261 lr=8.520302e-06 elapsed=10:39:31 avg_step=1.33s recent_step=1.46s eta=26:14:53 +2026-08-25 14:35:15 | INFO | train rank=0 | step=28890/100000 progress=28.9% loss=0.066315 lr=8.519128e-06 elapsed=10:39:45 avg_step=1.33s recent_step=1.43s eta=26:14:42 +2026-08-25 14:35:32 | INFO | train rank=0 | step=28900/100000 progress=28.9% loss=0.045151 lr=8.517953e-06 elapsed=10:40:01 avg_step=1.33s recent_step=1.62s eta=26:14:36 +2026-08-25 14:35:43 | INFO | train rank=0 | step=28910/100000 progress=28.9% loss=0.064911 lr=8.516778e-06 elapsed=10:40:13 avg_step=1.33s recent_step=1.12s eta=26:14:18 +2026-08-25 14:36:00 | INFO | train rank=0 | step=28920/100000 progress=28.9% loss=0.089172 lr=8.515602e-06 elapsed=10:40:30 avg_step=1.33s recent_step=1.76s eta=26:14:15 +2026-08-25 14:36:12 | INFO | train rank=0 | step=28930/100000 progress=28.9% loss=0.053223 lr=8.514426e-06 elapsed=10:40:42 avg_step=1.33s recent_step=1.15s eta=26:13:57 +2026-08-25 14:36:24 | INFO | train rank=0 | step=28940/100000 progress=28.9% loss=0.079651 lr=8.513250e-06 elapsed=10:40:54 avg_step=1.33s recent_step=1.18s eta=26:13:40 +2026-08-25 14:36:35 | INFO | train rank=0 | step=28950/100000 progress=28.9% loss=0.044754 lr=8.512073e-06 elapsed=10:41:05 avg_step=1.33s recent_step=1.14s eta=26:13:23 +2026-08-25 14:36:47 | INFO | train rank=0 | step=28960/100000 progress=29.0% loss=0.046539 lr=8.510896e-06 elapsed=10:41:16 avg_step=1.33s recent_step=1.15s eta=26:13:05 +2026-08-25 14:37:00 | INFO | train rank=0 | step=28970/100000 progress=29.0% loss=0.078278 lr=8.509719e-06 elapsed=10:41:30 avg_step=1.33s recent_step=1.31s eta=26:12:51 +2026-08-25 14:37:17 | INFO | train rank=0 | step=28980/100000 progress=29.0% loss=0.049988 lr=8.508541e-06 elapsed=10:41:47 avg_step=1.33s recent_step=1.78s eta=26:12:49 +2026-08-25 14:37:29 | INFO | train rank=0 | step=28990/100000 progress=29.0% loss=0.056641 lr=8.507363e-06 elapsed=10:41:59 avg_step=1.33s recent_step=1.16s eta=26:12:32 +2026-08-25 14:37:43 | INFO | train rank=0 | step=29000/100000 progress=29.0% loss=0.039246 lr=8.506184e-06 elapsed=10:42:13 avg_step=1.33s recent_step=1.41s eta=26:12:20 +2026-08-25 14:37:55 | INFO | train rank=0 | step=29010/100000 progress=29.0% loss=0.043137 lr=8.505005e-06 elapsed=10:42:24 avg_step=1.33s recent_step=1.14s eta=26:12:02 +2026-08-25 14:38:06 | INFO | train rank=0 | step=29020/100000 progress=29.0% loss=0.069366 lr=8.503826e-06 elapsed=10:42:36 avg_step=1.33s recent_step=1.16s eta=26:11:45 +2026-08-25 14:38:23 | INFO | train rank=0 | step=29030/100000 progress=29.0% loss=0.087372 lr=8.502646e-06 elapsed=10:42:52 avg_step=1.33s recent_step=1.64s eta=26:11:39 +2026-08-25 14:38:38 | INFO | train rank=0 | step=29040/100000 progress=29.0% loss=0.104675 lr=8.501466e-06 elapsed=10:43:08 avg_step=1.33s recent_step=1.54s eta=26:11:31 +2026-08-25 14:38:50 | INFO | train rank=0 | step=29050/100000 progress=29.0% loss=0.041992 lr=8.500285e-06 elapsed=10:43:20 avg_step=1.33s recent_step=1.18s eta=26:11:14 +2026-08-25 14:39:07 | INFO | train rank=0 | step=29060/100000 progress=29.1% loss=0.098816 lr=8.499104e-06 elapsed=10:43:37 avg_step=1.33s recent_step=1.71s eta=26:11:10 +2026-08-25 14:39:20 | INFO | train rank=0 | step=29070/100000 progress=29.1% loss=0.057770 lr=8.497923e-06 elapsed=10:43:50 avg_step=1.33s recent_step=1.31s eta=26:10:56 +2026-08-25 14:39:31 | INFO | train rank=0 | step=29080/100000 progress=29.1% loss=0.104431 lr=8.496741e-06 elapsed=10:44:01 avg_step=1.33s recent_step=1.15s eta=26:10:39 +2026-08-25 14:39:45 | INFO | train rank=0 | step=29090/100000 progress=29.1% loss=0.064575 lr=8.495559e-06 elapsed=10:44:15 avg_step=1.33s recent_step=1.33s eta=26:10:26 +2026-08-25 14:39:56 | INFO | train rank=0 | step=29100/100000 progress=29.1% loss=0.072723 lr=8.494377e-06 elapsed=10:44:26 avg_step=1.33s recent_step=1.14s eta=26:10:08 +2026-08-25 14:40:10 | INFO | train rank=0 | step=29110/100000 progress=29.1% loss=0.057129 lr=8.493194e-06 elapsed=10:44:40 avg_step=1.33s recent_step=1.40s eta=26:09:56 +2026-08-25 14:40:27 | INFO | train rank=0 | step=29120/100000 progress=29.1% loss=0.091064 lr=8.492011e-06 elapsed=10:44:57 avg_step=1.33s recent_step=1.69s eta=26:09:51 +2026-08-25 14:40:39 | INFO | train rank=0 | step=29130/100000 progress=29.1% loss=0.061951 lr=8.490827e-06 elapsed=10:45:08 avg_step=1.33s recent_step=1.16s eta=26:09:34 +2026-08-25 14:40:50 | INFO | train rank=0 | step=29140/100000 progress=29.1% loss=0.054047 lr=8.489643e-06 elapsed=10:45:20 avg_step=1.33s recent_step=1.13s eta=26:09:16 +2026-08-25 14:41:02 | INFO | train rank=0 | step=29150/100000 progress=29.1% loss=0.067169 lr=8.488459e-06 elapsed=10:45:32 avg_step=1.33s recent_step=1.20s eta=26:08:59 +2026-08-25 14:41:13 | INFO | train rank=0 | step=29160/100000 progress=29.2% loss=0.150635 lr=8.487274e-06 elapsed=10:45:43 avg_step=1.33s recent_step=1.14s eta=26:08:42 +2026-08-25 14:41:28 | INFO | train rank=0 | step=29170/100000 progress=29.2% loss=0.069733 lr=8.486089e-06 elapsed=10:45:58 avg_step=1.33s recent_step=1.45s eta=26:08:31 +2026-08-25 14:41:42 | INFO | train rank=0 | step=29180/100000 progress=29.2% loss=0.066803 lr=8.484903e-06 elapsed=10:46:12 avg_step=1.33s recent_step=1.48s eta=26:08:21 +2026-08-25 14:41:56 | INFO | train rank=0 | step=29190/100000 progress=29.2% loss=0.049927 lr=8.483718e-06 elapsed=10:46:26 avg_step=1.33s recent_step=1.40s eta=26:08:10 +2026-08-25 14:42:08 | INFO | train rank=0 | step=29200/100000 progress=29.2% loss=0.043137 lr=8.482531e-06 elapsed=10:46:38 avg_step=1.33s recent_step=1.17s eta=26:07:53 +2026-08-25 14:42:24 | INFO | train rank=0 | step=29210/100000 progress=29.2% loss=0.053467 lr=8.481345e-06 elapsed=10:46:54 avg_step=1.33s recent_step=1.56s eta=26:07:45 +2026-08-25 14:42:35 | INFO | train rank=0 | step=29220/100000 progress=29.2% loss=0.058716 lr=8.480158e-06 elapsed=10:47:05 avg_step=1.33s recent_step=1.10s eta=26:07:26 +2026-08-25 14:42:48 | INFO | train rank=0 | step=29230/100000 progress=29.2% loss=0.046677 lr=8.478970e-06 elapsed=10:47:18 avg_step=1.33s recent_step=1.37s eta=26:07:14 +2026-08-25 14:43:02 | INFO | train rank=0 | step=29240/100000 progress=29.2% loss=0.079956 lr=8.477782e-06 elapsed=10:47:32 avg_step=1.33s recent_step=1.38s eta=26:07:02 +2026-08-25 14:43:19 | INFO | train rank=0 | step=29250/100000 progress=29.2% loss=0.073395 lr=8.476594e-06 elapsed=10:47:49 avg_step=1.33s recent_step=1.71s eta=26:06:58 +2026-08-25 14:43:34 | INFO | train rank=0 | step=29260/100000 progress=29.3% loss=0.065277 lr=8.475406e-06 elapsed=10:48:04 avg_step=1.33s recent_step=1.44s eta=26:06:48 +2026-08-25 14:43:45 | INFO | train rank=0 | step=29270/100000 progress=29.3% loss=0.062286 lr=8.474217e-06 elapsed=10:48:15 avg_step=1.33s recent_step=1.13s eta=26:06:30 +2026-08-25 14:43:57 | INFO | train rank=0 | step=29280/100000 progress=29.3% loss=0.056870 lr=8.473028e-06 elapsed=10:48:27 avg_step=1.33s recent_step=1.20s eta=26:06:13 +2026-08-25 14:44:12 | INFO | train rank=0 | step=29290/100000 progress=29.3% loss=0.054794 lr=8.471838e-06 elapsed=10:48:42 avg_step=1.33s recent_step=1.46s eta=26:06:03 +2026-08-25 14:44:24 | INFO | train rank=0 | step=29300/100000 progress=29.3% loss=0.056976 lr=8.470648e-06 elapsed=10:48:54 avg_step=1.33s recent_step=1.19s eta=26:05:46 +2026-08-25 14:44:35 | INFO | train rank=0 | step=29310/100000 progress=29.3% loss=0.090881 lr=8.469457e-06 elapsed=10:49:05 avg_step=1.33s recent_step=1.13s eta=26:05:28 +2026-08-25 14:44:49 | INFO | train rank=0 | step=29320/100000 progress=29.3% loss=0.063263 lr=8.468267e-06 elapsed=10:49:19 avg_step=1.33s recent_step=1.41s eta=26:05:17 +2026-08-25 14:45:01 | INFO | train rank=0 | step=29330/100000 progress=29.3% loss=0.098816 lr=8.467075e-06 elapsed=10:49:31 avg_step=1.33s recent_step=1.16s eta=26:05:00 +2026-08-25 14:45:12 | INFO | train rank=0 | step=29340/100000 progress=29.3% loss=0.132202 lr=8.465884e-06 elapsed=10:49:42 avg_step=1.33s recent_step=1.15s eta=26:04:42 +2026-08-25 14:45:24 | INFO | train rank=0 | step=29350/100000 progress=29.3% loss=0.042595 lr=8.464692e-06 elapsed=10:49:54 avg_step=1.33s recent_step=1.16s eta=26:04:25 +2026-08-25 14:45:38 | INFO | train rank=0 | step=29360/100000 progress=29.4% loss=0.047226 lr=8.463499e-06 elapsed=10:50:08 avg_step=1.33s recent_step=1.37s eta=26:04:13 +2026-08-25 14:45:49 | INFO | train rank=0 | step=29370/100000 progress=29.4% loss=0.071655 lr=8.462307e-06 elapsed=10:50:19 avg_step=1.33s recent_step=1.17s eta=26:03:56 +2026-08-25 14:46:01 | INFO | train rank=0 | step=29380/100000 progress=29.4% loss=0.116577 lr=8.461114e-06 elapsed=10:50:31 avg_step=1.33s recent_step=1.15s eta=26:03:38 +2026-08-25 14:46:15 | INFO | train rank=0 | step=29390/100000 progress=29.4% loss=0.053436 lr=8.459920e-06 elapsed=10:50:45 avg_step=1.33s recent_step=1.44s eta=26:03:27 +2026-08-25 14:46:27 | INFO | train rank=0 | step=29400/100000 progress=29.4% loss=0.054436 lr=8.458726e-06 elapsed=10:50:57 avg_step=1.33s recent_step=1.21s eta=26:03:11 +2026-08-25 14:46:39 | INFO | train rank=0 | step=29410/100000 progress=29.4% loss=0.057587 lr=8.457532e-06 elapsed=10:51:09 avg_step=1.33s recent_step=1.18s eta=26:02:54 +2026-08-25 14:46:52 | INFO | train rank=0 | step=29420/100000 progress=29.4% loss=0.118164 lr=8.456337e-06 elapsed=10:51:22 avg_step=1.33s recent_step=1.30s eta=26:02:40 +2026-08-25 14:47:06 | INFO | train rank=0 | step=29430/100000 progress=29.4% loss=0.042953 lr=8.455143e-06 elapsed=10:51:36 avg_step=1.33s recent_step=1.40s eta=26:02:29 +2026-08-25 14:47:20 | INFO | train rank=0 | step=29440/100000 progress=29.4% loss=0.042862 lr=8.453947e-06 elapsed=10:51:50 avg_step=1.33s recent_step=1.35s eta=26:02:16 +2026-08-25 14:47:33 | INFO | train rank=0 | step=29450/100000 progress=29.4% loss=0.062378 lr=8.452751e-06 elapsed=10:52:03 avg_step=1.33s recent_step=1.34s eta=26:02:03 +2026-08-25 14:47:46 | INFO | train rank=0 | step=29460/100000 progress=29.5% loss=0.046280 lr=8.451555e-06 elapsed=10:52:15 avg_step=1.33s recent_step=1.25s eta=26:01:48 +2026-08-25 14:48:00 | INFO | train rank=0 | step=29470/100000 progress=29.5% loss=0.056793 lr=8.450359e-06 elapsed=10:52:29 avg_step=1.33s recent_step=1.40s eta=26:01:36 +2026-08-25 14:48:11 | INFO | train rank=0 | step=29480/100000 progress=29.5% loss=0.057251 lr=8.449162e-06 elapsed=10:52:41 avg_step=1.33s recent_step=1.20s eta=26:01:20 +2026-08-25 14:48:23 | INFO | train rank=0 | step=29490/100000 progress=29.5% loss=0.072723 lr=8.447965e-06 elapsed=10:52:53 avg_step=1.33s recent_step=1.17s eta=26:01:03 +2026-08-25 14:48:39 | INFO | train rank=0 | step=29500/100000 progress=29.5% loss=0.049561 lr=8.446767e-06 elapsed=10:53:09 avg_step=1.33s recent_step=1.58s eta=26:00:56 +2026-08-25 14:48:52 | INFO | train rank=0 | step=29510/100000 progress=29.5% loss=0.047424 lr=8.445569e-06 elapsed=10:53:22 avg_step=1.33s recent_step=1.34s eta=26:00:43 +2026-08-25 14:49:04 | INFO | train rank=0 | step=29520/100000 progress=29.5% loss=0.034714 lr=8.444371e-06 elapsed=10:53:34 avg_step=1.33s recent_step=1.19s eta=26:00:26 +2026-08-25 14:49:16 | INFO | train rank=0 | step=29530/100000 progress=29.5% loss=0.075745 lr=8.443172e-06 elapsed=10:53:46 avg_step=1.33s recent_step=1.20s eta=26:00:10 +2026-08-25 14:49:32 | INFO | train rank=0 | step=29540/100000 progress=29.5% loss=0.069092 lr=8.441973e-06 elapsed=10:54:02 avg_step=1.33s recent_step=1.54s eta=26:00:01 +2026-08-25 14:49:49 | INFO | train rank=0 | step=29550/100000 progress=29.5% loss=0.148254 lr=8.440773e-06 elapsed=10:54:19 avg_step=1.33s recent_step=1.75s eta=25:59:58 +2026-08-25 14:50:01 | INFO | train rank=0 | step=29560/100000 progress=29.6% loss=0.056931 lr=8.439573e-06 elapsed=10:54:31 avg_step=1.33s recent_step=1.15s eta=25:59:40 +2026-08-25 14:50:15 | INFO | train rank=0 | step=29570/100000 progress=29.6% loss=0.066986 lr=8.438373e-06 elapsed=10:54:45 avg_step=1.33s recent_step=1.42s eta=25:59:29 +2026-08-25 14:50:27 | INFO | train rank=0 | step=29580/100000 progress=29.6% loss=0.057159 lr=8.437172e-06 elapsed=10:54:56 avg_step=1.33s recent_step=1.17s eta=25:59:12 +2026-08-25 14:50:42 | INFO | train rank=0 | step=29590/100000 progress=29.6% loss=0.055573 lr=8.435971e-06 elapsed=10:55:12 avg_step=1.33s recent_step=1.54s eta=25:59:04 +2026-08-25 14:50:56 | INFO | train rank=0 | step=29600/100000 progress=29.6% loss=0.084412 lr=8.434770e-06 elapsed=10:55:26 avg_step=1.33s recent_step=1.43s eta=25:58:53 +2026-08-25 14:51:11 | INFO | train rank=0 | step=29610/100000 progress=29.6% loss=0.075653 lr=8.433568e-06 elapsed=10:55:41 avg_step=1.33s recent_step=1.51s eta=25:58:44 +2026-08-25 14:51:23 | INFO | train rank=0 | step=29620/100000 progress=29.6% loss=0.035126 lr=8.432366e-06 elapsed=10:55:53 avg_step=1.33s recent_step=1.16s eta=25:58:27 +2026-08-25 14:51:37 | INFO | train rank=0 | step=29630/100000 progress=29.6% loss=0.034653 lr=8.431164e-06 elapsed=10:56:07 avg_step=1.33s recent_step=1.37s eta=25:58:15 +2026-08-25 14:51:53 | INFO | train rank=0 | step=29640/100000 progress=29.6% loss=0.040573 lr=8.429961e-06 elapsed=10:56:23 avg_step=1.33s recent_step=1.63s eta=25:58:09 +2026-08-25 14:52:05 | INFO | train rank=0 | step=29650/100000 progress=29.6% loss=0.062775 lr=8.428757e-06 elapsed=10:56:35 avg_step=1.33s recent_step=1.18s eta=25:57:52 +2026-08-25 14:52:16 | INFO | train rank=0 | step=29660/100000 progress=29.7% loss=0.048752 lr=8.427554e-06 elapsed=10:56:46 avg_step=1.33s recent_step=1.16s eta=25:57:34 +2026-08-25 14:52:31 | INFO | train rank=0 | step=29670/100000 progress=29.7% loss=0.035294 lr=8.426350e-06 elapsed=10:57:01 avg_step=1.33s recent_step=1.50s eta=25:57:25 +2026-08-25 14:52:45 | INFO | train rank=0 | step=29680/100000 progress=29.7% loss=0.073181 lr=8.425145e-06 elapsed=10:57:15 avg_step=1.33s recent_step=1.35s eta=25:57:12 +2026-08-25 14:52:56 | INFO | train rank=0 | step=29690/100000 progress=29.7% loss=0.083618 lr=8.423941e-06 elapsed=10:57:26 avg_step=1.33s recent_step=1.15s eta=25:56:55 +2026-08-25 14:53:08 | INFO | train rank=0 | step=29700/100000 progress=29.7% loss=0.051224 lr=8.422736e-06 elapsed=10:57:38 avg_step=1.33s recent_step=1.18s eta=25:56:38 +2026-08-25 14:53:28 | INFO | train rank=0 | step=29710/100000 progress=29.7% loss=0.055054 lr=8.421530e-06 elapsed=10:57:58 avg_step=1.33s recent_step=2.00s eta=25:56:41 +2026-08-25 14:53:41 | INFO | train rank=0 | step=29720/100000 progress=29.7% loss=0.048035 lr=8.420324e-06 elapsed=10:58:11 avg_step=1.33s recent_step=1.28s eta=25:56:26 +2026-08-25 14:53:53 | INFO | train rank=0 | step=29730/100000 progress=29.7% loss=0.063446 lr=8.419118e-06 elapsed=10:58:23 avg_step=1.33s recent_step=1.17s eta=25:56:09 +2026-08-25 14:54:04 | INFO | train rank=0 | step=29740/100000 progress=29.7% loss=0.051910 lr=8.417911e-06 elapsed=10:58:34 avg_step=1.33s recent_step=1.12s eta=25:55:51 +2026-08-25 14:54:19 | INFO | train rank=0 | step=29750/100000 progress=29.8% loss=0.067688 lr=8.416704e-06 elapsed=10:58:48 avg_step=1.33s recent_step=1.48s eta=25:55:41 +2026-08-25 14:54:31 | INFO | train rank=0 | step=29760/100000 progress=29.8% loss=0.052963 lr=8.415497e-06 elapsed=10:59:01 avg_step=1.33s recent_step=1.21s eta=25:55:25 +2026-08-25 14:54:46 | INFO | train rank=0 | step=29770/100000 progress=29.8% loss=0.063873 lr=8.414289e-06 elapsed=10:59:16 avg_step=1.33s recent_step=1.50s eta=25:55:16 +2026-08-25 14:54:57 | INFO | train rank=0 | step=29780/100000 progress=29.8% loss=0.046829 lr=8.413081e-06 elapsed=10:59:27 avg_step=1.33s recent_step=1.13s eta=25:54:58 +2026-08-25 14:55:17 | INFO | train rank=0 | step=29790/100000 progress=29.8% loss=0.042770 lr=8.411872e-06 elapsed=10:59:47 avg_step=1.33s recent_step=1.99s eta=25:55:00 +2026-08-25 14:55:29 | INFO | train rank=0 | step=29800/100000 progress=29.8% loss=0.070953 lr=8.410664e-06 elapsed=10:59:59 avg_step=1.33s recent_step=1.18s eta=25:54:43 +2026-08-25 14:55:41 | INFO | train rank=0 | step=29810/100000 progress=29.8% loss=0.059204 lr=8.409454e-06 elapsed=11:00:11 avg_step=1.33s recent_step=1.22s eta=25:54:27 +2026-08-25 14:55:52 | INFO | train rank=0 | step=29820/100000 progress=29.8% loss=0.065536 lr=8.408245e-06 elapsed=11:00:22 avg_step=1.33s recent_step=1.14s eta=25:54:09 +2026-08-25 14:56:04 | INFO | train rank=0 | step=29830/100000 progress=29.8% loss=0.037460 lr=8.407035e-06 elapsed=11:00:34 avg_step=1.33s recent_step=1.15s eta=25:53:52 +2026-08-25 14:56:15 | INFO | train rank=0 | step=29840/100000 progress=29.8% loss=0.043228 lr=8.405824e-06 elapsed=11:00:45 avg_step=1.33s recent_step=1.18s eta=25:53:35 +2026-08-25 14:56:29 | INFO | train rank=0 | step=29850/100000 progress=29.8% loss=0.060913 lr=8.404614e-06 elapsed=11:00:59 avg_step=1.33s recent_step=1.34s eta=25:53:22 +2026-08-25 14:56:44 | INFO | train rank=0 | step=29860/100000 progress=29.9% loss=0.058243 lr=8.403402e-06 elapsed=11:01:14 avg_step=1.33s recent_step=1.51s eta=25:53:13 +2026-08-25 14:56:56 | INFO | train rank=0 | step=29870/100000 progress=29.9% loss=0.061523 lr=8.402191e-06 elapsed=11:01:26 avg_step=1.33s recent_step=1.20s eta=25:52:57 +2026-08-25 14:57:10 | INFO | train rank=0 | step=29880/100000 progress=29.9% loss=0.059326 lr=8.400979e-06 elapsed=11:01:40 avg_step=1.33s recent_step=1.38s eta=25:52:45 +2026-08-25 14:57:24 | INFO | train rank=0 | step=29890/100000 progress=29.9% loss=0.037819 lr=8.399767e-06 elapsed=11:01:54 avg_step=1.33s recent_step=1.46s eta=25:52:35 +2026-08-25 14:57:36 | INFO | train rank=0 | step=29900/100000 progress=29.9% loss=0.068634 lr=8.398554e-06 elapsed=11:02:06 avg_step=1.33s recent_step=1.13s eta=25:52:17 +2026-08-25 14:57:51 | INFO | train rank=0 | step=29910/100000 progress=29.9% loss=0.049454 lr=8.397341e-06 elapsed=11:02:21 avg_step=1.33s recent_step=1.57s eta=25:52:09 +2026-08-25 14:58:05 | INFO | train rank=0 | step=29920/100000 progress=29.9% loss=0.055237 lr=8.396128e-06 elapsed=11:02:35 avg_step=1.33s recent_step=1.36s eta=25:51:57 +2026-08-25 14:58:19 | INFO | train rank=0 | step=29930/100000 progress=29.9% loss=0.050339 lr=8.394914e-06 elapsed=11:02:49 avg_step=1.33s recent_step=1.40s eta=25:51:45 +2026-08-25 14:58:34 | INFO | train rank=0 | step=29940/100000 progress=29.9% loss=0.053391 lr=8.393700e-06 elapsed=11:03:04 avg_step=1.33s recent_step=1.53s eta=25:51:36 +2026-08-25 14:58:49 | INFO | train rank=0 | step=29950/100000 progress=29.9% loss=0.041367 lr=8.392486e-06 elapsed=11:03:19 avg_step=1.33s recent_step=1.44s eta=25:51:26 +2026-08-25 14:59:01 | INFO | train rank=0 | step=29960/100000 progress=30.0% loss=0.040787 lr=8.391271e-06 elapsed=11:03:31 avg_step=1.33s recent_step=1.26s eta=25:51:11 +2026-08-25 14:59:13 | INFO | train rank=0 | step=29970/100000 progress=30.0% loss=0.074417 lr=8.390056e-06 elapsed=11:03:43 avg_step=1.33s recent_step=1.20s eta=25:50:55 +2026-08-25 14:59:26 | INFO | train rank=0 | step=29980/100000 progress=30.0% loss=0.081543 lr=8.388840e-06 elapsed=11:03:55 avg_step=1.33s recent_step=1.21s eta=25:50:39 +2026-08-25 14:59:40 | INFO | train rank=0 | step=29990/100000 progress=30.0% loss=0.041245 lr=8.387624e-06 elapsed=11:04:10 avg_step=1.33s recent_step=1.47s eta=25:50:29 +2026-08-25 14:59:55 | INFO | train rank=0 | step=30000/100000 progress=30.0% loss=0.045410 lr=8.386408e-06 elapsed=11:04:25 avg_step=1.33s recent_step=1.51s eta=25:50:19 +2026-08-25 15:00:09 | INFO | train rank=0 | step=30010/100000 progress=30.0% loss=0.128967 lr=8.385191e-06 elapsed=11:04:39 avg_step=1.33s recent_step=1.35s eta=25:50:07 +2026-08-25 15:00:21 | INFO | train rank=0 | step=30020/100000 progress=30.0% loss=0.050232 lr=8.383974e-06 elapsed=11:04:51 avg_step=1.33s recent_step=1.25s eta=25:49:51 +2026-08-25 15:00:34 | INFO | train rank=0 | step=30030/100000 progress=30.0% loss=0.075226 lr=8.382757e-06 elapsed=11:05:04 avg_step=1.33s recent_step=1.26s eta=25:49:37 +2026-08-25 15:00:47 | INFO | train rank=0 | step=30040/100000 progress=30.0% loss=0.045219 lr=8.381539e-06 elapsed=11:05:17 avg_step=1.33s recent_step=1.30s eta=25:49:23 +2026-08-25 15:01:01 | INFO | train rank=0 | step=30050/100000 progress=30.0% loss=0.083130 lr=8.380321e-06 elapsed=11:05:31 avg_step=1.33s recent_step=1.43s eta=25:49:12 +2026-08-25 15:01:20 | INFO | train rank=0 | step=30060/100000 progress=30.1% loss=0.048477 lr=8.379102e-06 elapsed=11:05:50 avg_step=1.33s recent_step=1.88s eta=25:49:11 +2026-08-25 15:01:31 | INFO | train rank=0 | step=30070/100000 progress=30.1% loss=0.068085 lr=8.377883e-06 elapsed=11:06:01 avg_step=1.33s recent_step=1.13s eta=25:48:53 +2026-08-25 15:01:43 | INFO | train rank=0 | step=30080/100000 progress=30.1% loss=0.074982 lr=8.376664e-06 elapsed=11:06:13 avg_step=1.33s recent_step=1.19s eta=25:48:37 +2026-08-25 15:01:55 | INFO | train rank=0 | step=30090/100000 progress=30.1% loss=0.051407 lr=8.375444e-06 elapsed=11:06:24 avg_step=1.33s recent_step=1.13s eta=25:48:19 +2026-08-25 15:02:06 | INFO | train rank=0 | step=30100/100000 progress=30.1% loss=0.056458 lr=8.374224e-06 elapsed=11:06:36 avg_step=1.33s recent_step=1.16s eta=25:48:02 +2026-08-25 15:02:18 | INFO | train rank=0 | step=30110/100000 progress=30.1% loss=0.064362 lr=8.373004e-06 elapsed=11:06:48 avg_step=1.33s recent_step=1.17s eta=25:47:45 +2026-08-25 15:02:32 | INFO | train rank=0 | step=30120/100000 progress=30.1% loss=0.068329 lr=8.371783e-06 elapsed=11:07:02 avg_step=1.33s recent_step=1.39s eta=25:47:33 +2026-08-25 15:02:47 | INFO | train rank=0 | step=30130/100000 progress=30.1% loss=0.041290 lr=8.370562e-06 elapsed=11:07:16 avg_step=1.33s recent_step=1.48s eta=25:47:23 +2026-08-25 15:03:00 | INFO | train rank=0 | step=30140/100000 progress=30.1% loss=0.065750 lr=8.369341e-06 elapsed=11:07:30 avg_step=1.33s recent_step=1.38s eta=25:47:11 +2026-08-25 15:03:12 | INFO | train rank=0 | step=30150/100000 progress=30.1% loss=0.044624 lr=8.368119e-06 elapsed=11:07:42 avg_step=1.33s recent_step=1.19s eta=25:46:55 +2026-08-25 15:03:26 | INFO | train rank=0 | step=30160/100000 progress=30.2% loss=0.082947 lr=8.366897e-06 elapsed=11:07:55 avg_step=1.33s recent_step=1.33s eta=25:46:41 +2026-08-25 15:03:37 | INFO | train rank=0 | step=30170/100000 progress=30.2% loss=0.055023 lr=8.365674e-06 elapsed=11:08:07 avg_step=1.33s recent_step=1.13s eta=25:46:23 +2026-08-25 15:03:54 | INFO | train rank=0 | step=30180/100000 progress=30.2% loss=0.073517 lr=8.364451e-06 elapsed=11:08:24 avg_step=1.33s recent_step=1.75s eta=25:46:20 +2026-08-25 15:04:06 | INFO | train rank=0 | step=30190/100000 progress=30.2% loss=0.059845 lr=8.363228e-06 elapsed=11:08:36 avg_step=1.33s recent_step=1.17s eta=25:46:03 +2026-08-25 15:04:21 | INFO | train rank=0 | step=30200/100000 progress=30.2% loss=0.037521 lr=8.362004e-06 elapsed=11:08:51 avg_step=1.33s recent_step=1.48s eta=25:45:53 +2026-08-25 15:04:33 | INFO | train rank=0 | step=30210/100000 progress=30.2% loss=0.036514 lr=8.360780e-06 elapsed=11:09:03 avg_step=1.33s recent_step=1.23s eta=25:45:37 +2026-08-25 15:04:46 | INFO | train rank=0 | step=30220/100000 progress=30.2% loss=0.059998 lr=8.359556e-06 elapsed=11:09:16 avg_step=1.33s recent_step=1.32s eta=25:45:24 +2026-08-25 15:05:00 | INFO | train rank=0 | step=30230/100000 progress=30.2% loss=0.040634 lr=8.358331e-06 elapsed=11:09:30 avg_step=1.33s recent_step=1.35s eta=25:45:11 +2026-08-25 15:05:12 | INFO | train rank=0 | step=30240/100000 progress=30.2% loss=0.036453 lr=8.357106e-06 elapsed=11:09:41 avg_step=1.33s recent_step=1.18s eta=25:44:54 +2026-08-25 15:05:27 | INFO | train rank=0 | step=30250/100000 progress=30.2% loss=0.063721 lr=8.355880e-06 elapsed=11:09:56 avg_step=1.33s recent_step=1.51s eta=25:44:45 +2026-08-25 15:05:41 | INFO | train rank=0 | step=30260/100000 progress=30.3% loss=0.048599 lr=8.354654e-06 elapsed=11:10:11 avg_step=1.33s recent_step=1.48s eta=25:44:35 +2026-08-25 15:05:53 | INFO | train rank=0 | step=30270/100000 progress=30.3% loss=0.048904 lr=8.353428e-06 elapsed=11:10:23 avg_step=1.33s recent_step=1.13s eta=25:44:18 +2026-08-25 15:06:07 | INFO | train rank=0 | step=30280/100000 progress=30.3% loss=0.060379 lr=8.352201e-06 elapsed=11:10:37 avg_step=1.33s recent_step=1.47s eta=25:44:07 +2026-08-25 15:06:20 | INFO | train rank=0 | step=30290/100000 progress=30.3% loss=0.048828 lr=8.350974e-06 elapsed=11:10:50 avg_step=1.33s recent_step=1.22s eta=25:43:52 +2026-08-25 15:06:33 | INFO | train rank=0 | step=30300/100000 progress=30.3% loss=0.055862 lr=8.349747e-06 elapsed=11:11:03 avg_step=1.33s recent_step=1.31s eta=25:43:38 +2026-08-25 15:06:45 | INFO | train rank=0 | step=30310/100000 progress=30.3% loss=0.055145 lr=8.348519e-06 elapsed=11:11:14 avg_step=1.33s recent_step=1.18s eta=25:43:21 +2026-08-25 15:06:59 | INFO | train rank=0 | step=30320/100000 progress=30.3% loss=0.068146 lr=8.347291e-06 elapsed=11:11:29 avg_step=1.33s recent_step=1.42s eta=25:43:10 +2026-08-25 15:07:10 | INFO | train rank=0 | step=30330/100000 progress=30.3% loss=0.083527 lr=8.346063e-06 elapsed=11:11:40 avg_step=1.33s recent_step=1.12s eta=25:42:52 +2026-08-25 15:07:22 | INFO | train rank=0 | step=30340/100000 progress=30.3% loss=0.048706 lr=8.344834e-06 elapsed=11:11:51 avg_step=1.33s recent_step=1.15s eta=25:42:35 +2026-08-25 15:07:33 | INFO | train rank=0 | step=30350/100000 progress=30.3% loss=0.068939 lr=8.343605e-06 elapsed=11:12:03 avg_step=1.33s recent_step=1.12s eta=25:42:17 +2026-08-25 15:07:48 | INFO | train rank=0 | step=30360/100000 progress=30.4% loss=0.060455 lr=8.342375e-06 elapsed=11:12:18 avg_step=1.33s recent_step=1.58s eta=25:42:09 +2026-08-25 15:08:00 | INFO | train rank=0 | step=30370/100000 progress=30.4% loss=0.046906 lr=8.341145e-06 elapsed=11:12:30 avg_step=1.33s recent_step=1.19s eta=25:41:53 +2026-08-25 15:08:14 | INFO | train rank=0 | step=30380/100000 progress=30.4% loss=0.043640 lr=8.339915e-06 elapsed=11:12:44 avg_step=1.33s recent_step=1.39s eta=25:41:41 +2026-08-25 15:08:31 | INFO | train rank=0 | step=30390/100000 progress=30.4% loss=0.046936 lr=8.338684e-06 elapsed=11:13:01 avg_step=1.33s recent_step=1.68s eta=25:41:36 +2026-08-25 15:08:45 | INFO | train rank=0 | step=30400/100000 progress=30.4% loss=0.058731 lr=8.337453e-06 elapsed=11:13:15 avg_step=1.33s recent_step=1.37s eta=25:41:23 +2026-08-25 15:08:56 | INFO | train rank=0 | step=30410/100000 progress=30.4% loss=0.060410 lr=8.336222e-06 elapsed=11:13:26 avg_step=1.33s recent_step=1.08s eta=25:41:04 +2026-08-25 15:09:08 | INFO | train rank=0 | step=30420/100000 progress=30.4% loss=0.074158 lr=8.334990e-06 elapsed=11:13:38 avg_step=1.33s recent_step=1.23s eta=25:40:49 +2026-08-25 15:09:19 | INFO | train rank=0 | step=30430/100000 progress=30.4% loss=0.046219 lr=8.333758e-06 elapsed=11:13:49 avg_step=1.33s recent_step=1.14s eta=25:40:31 +2026-08-25 15:09:32 | INFO | train rank=0 | step=30440/100000 progress=30.4% loss=0.032867 lr=8.332525e-06 elapsed=11:14:02 avg_step=1.33s recent_step=1.30s eta=25:40:17 +2026-08-25 15:09:46 | INFO | train rank=0 | step=30450/100000 progress=30.4% loss=0.053833 lr=8.331293e-06 elapsed=11:14:16 avg_step=1.33s recent_step=1.38s eta=25:40:05 +2026-08-25 15:09:57 | INFO | train rank=0 | step=30460/100000 progress=30.5% loss=0.047180 lr=8.330059e-06 elapsed=11:14:27 avg_step=1.33s recent_step=1.11s eta=25:39:47 +2026-08-25 15:10:16 | INFO | train rank=0 | step=30470/100000 progress=30.5% loss=0.046707 lr=8.328826e-06 elapsed=11:14:46 avg_step=1.33s recent_step=1.85s eta=25:39:46 +2026-08-25 15:10:28 | INFO | train rank=0 | step=30480/100000 progress=30.5% loss=0.053925 lr=8.327592e-06 elapsed=11:14:58 avg_step=1.33s recent_step=1.18s eta=25:39:29 +2026-08-25 15:10:39 | INFO | train rank=0 | step=30490/100000 progress=30.5% loss=0.053223 lr=8.326358e-06 elapsed=11:15:08 avg_step=1.33s recent_step=1.09s eta=25:39:10 +2026-08-25 15:10:52 | INFO | train rank=0 | step=30500/100000 progress=30.5% loss=0.071533 lr=8.325123e-06 elapsed=11:15:22 avg_step=1.33s recent_step=1.39s eta=25:38:59 +2026-08-25 15:11:05 | INFO | train rank=0 | step=30510/100000 progress=30.5% loss=0.046234 lr=8.323888e-06 elapsed=11:15:35 avg_step=1.33s recent_step=1.29s eta=25:38:44 +2026-08-25 15:11:22 | INFO | train rank=0 | step=30520/100000 progress=30.5% loss=0.063232 lr=8.322652e-06 elapsed=11:15:52 avg_step=1.33s recent_step=1.64s eta=25:38:38 +2026-08-25 15:11:33 | INFO | train rank=0 | step=30530/100000 progress=30.5% loss=0.090790 lr=8.321417e-06 elapsed=11:16:03 avg_step=1.33s recent_step=1.16s eta=25:38:21 +2026-08-25 15:11:46 | INFO | train rank=0 | step=30540/100000 progress=30.5% loss=0.060577 lr=8.320181e-06 elapsed=11:16:16 avg_step=1.33s recent_step=1.29s eta=25:38:07 +2026-08-25 15:12:01 | INFO | train rank=0 | step=30550/100000 progress=30.6% loss=0.048126 lr=8.318944e-06 elapsed=11:16:30 avg_step=1.33s recent_step=1.43s eta=25:37:56 +2026-08-25 15:12:15 | INFO | train rank=0 | step=30560/100000 progress=30.6% loss=0.062408 lr=8.317707e-06 elapsed=11:16:45 avg_step=1.33s recent_step=1.44s eta=25:37:45 +2026-08-25 15:12:26 | INFO | train rank=0 | step=30570/100000 progress=30.6% loss=0.039688 lr=8.316470e-06 elapsed=11:16:56 avg_step=1.33s recent_step=1.10s eta=25:37:27 +2026-08-25 15:12:41 | INFO | train rank=0 | step=30580/100000 progress=30.6% loss=0.033607 lr=8.315232e-06 elapsed=11:17:11 avg_step=1.33s recent_step=1.53s eta=25:37:18 +2026-08-25 15:12:55 | INFO | train rank=0 | step=30590/100000 progress=30.6% loss=0.054840 lr=8.313995e-06 elapsed=11:17:25 avg_step=1.33s recent_step=1.40s eta=25:37:06 +2026-08-25 15:13:09 | INFO | train rank=0 | step=30600/100000 progress=30.6% loss=0.036896 lr=8.312756e-06 elapsed=11:17:39 avg_step=1.33s recent_step=1.40s eta=25:36:55 +2026-08-25 15:13:24 | INFO | train rank=0 | step=30610/100000 progress=30.6% loss=0.045128 lr=8.311518e-06 elapsed=11:17:53 avg_step=1.33s recent_step=1.42s eta=25:36:43 +2026-08-25 15:13:35 | INFO | train rank=0 | step=30620/100000 progress=30.6% loss=0.051514 lr=8.310279e-06 elapsed=11:18:05 avg_step=1.33s recent_step=1.17s eta=25:36:27 +2026-08-25 15:13:52 | INFO | train rank=0 | step=30630/100000 progress=30.6% loss=0.067703 lr=8.309039e-06 elapsed=11:18:22 avg_step=1.33s recent_step=1.66s eta=25:36:21 +2026-08-25 15:14:04 | INFO | train rank=0 | step=30640/100000 progress=30.6% loss=0.064529 lr=8.307799e-06 elapsed=11:18:34 avg_step=1.33s recent_step=1.20s eta=25:36:05 +2026-08-25 15:14:15 | INFO | train rank=0 | step=30650/100000 progress=30.6% loss=0.051666 lr=8.306559e-06 elapsed=11:18:45 avg_step=1.33s recent_step=1.11s eta=25:35:46 +2026-08-25 15:14:30 | INFO | train rank=0 | step=30660/100000 progress=30.7% loss=0.057251 lr=8.305319e-06 elapsed=11:19:00 avg_step=1.33s recent_step=1.52s eta=25:35:37 +2026-08-25 15:14:42 | INFO | train rank=0 | step=30670/100000 progress=30.7% loss=0.060852 lr=8.304078e-06 elapsed=11:19:12 avg_step=1.33s recent_step=1.23s eta=25:35:22 +2026-08-25 15:14:56 | INFO | train rank=0 | step=30680/100000 progress=30.7% loss=0.065613 lr=8.302837e-06 elapsed=11:19:26 avg_step=1.33s recent_step=1.36s eta=25:35:09 +2026-08-25 15:15:08 | INFO | train rank=0 | step=30690/100000 progress=30.7% loss=0.042694 lr=8.301595e-06 elapsed=11:19:37 avg_step=1.33s recent_step=1.15s eta=25:34:52 +2026-08-25 15:15:19 | INFO | train rank=0 | step=30700/100000 progress=30.7% loss=0.055511 lr=8.300353e-06 elapsed=11:19:48 avg_step=1.33s recent_step=1.10s eta=25:34:33 +2026-08-25 15:15:32 | INFO | train rank=0 | step=30710/100000 progress=30.7% loss=0.042313 lr=8.299111e-06 elapsed=11:20:02 avg_step=1.33s recent_step=1.34s eta=25:34:20 +2026-08-25 15:15:48 | INFO | train rank=0 | step=30720/100000 progress=30.7% loss=0.056793 lr=8.297868e-06 elapsed=11:20:18 avg_step=1.33s recent_step=1.59s eta=25:34:13 +2026-08-25 15:16:02 | INFO | train rank=0 | step=30730/100000 progress=30.7% loss=0.040009 lr=8.296625e-06 elapsed=11:20:31 avg_step=1.33s recent_step=1.38s eta=25:34:01 +2026-08-25 15:16:14 | INFO | train rank=0 | step=30740/100000 progress=30.7% loss=0.052277 lr=8.295382e-06 elapsed=11:20:44 avg_step=1.33s recent_step=1.26s eta=25:33:46 +2026-08-25 15:16:25 | INFO | train rank=0 | step=30750/100000 progress=30.8% loss=0.085022 lr=8.294138e-06 elapsed=11:20:55 avg_step=1.33s recent_step=1.11s eta=25:33:28 +2026-08-25 15:16:37 | INFO | train rank=0 | step=30760/100000 progress=30.8% loss=0.097229 lr=8.292894e-06 elapsed=11:21:07 avg_step=1.33s recent_step=1.17s eta=25:33:11 +2026-08-25 15:16:50 | INFO | train rank=0 | step=30770/100000 progress=30.8% loss=0.080414 lr=8.291650e-06 elapsed=11:21:20 avg_step=1.33s recent_step=1.36s eta=25:32:58 +2026-08-25 15:17:02 | INFO | train rank=0 | step=30780/100000 progress=30.8% loss=0.040329 lr=8.290405e-06 elapsed=11:21:32 avg_step=1.33s recent_step=1.14s eta=25:32:40 +2026-08-25 15:17:22 | INFO | train rank=0 | step=30790/100000 progress=30.8% loss=0.037788 lr=8.289160e-06 elapsed=11:21:52 avg_step=1.33s recent_step=2.06s eta=25:32:44 +2026-08-25 15:17:37 | INFO | train rank=0 | step=30800/100000 progress=30.8% loss=0.054993 lr=8.287914e-06 elapsed=11:22:07 avg_step=1.33s recent_step=1.47s eta=25:32:34 +2026-08-25 15:17:49 | INFO | train rank=0 | step=30810/100000 progress=30.8% loss=0.041000 lr=8.286669e-06 elapsed=11:22:19 avg_step=1.33s recent_step=1.16s eta=25:32:16 +2026-08-25 15:18:00 | INFO | train rank=0 | step=30820/100000 progress=30.8% loss=0.053375 lr=8.285422e-06 elapsed=11:22:30 avg_step=1.33s recent_step=1.16s eta=25:31:59 +2026-08-25 15:18:16 | INFO | train rank=0 | step=30830/100000 progress=30.8% loss=0.066193 lr=8.284176e-06 elapsed=11:22:45 avg_step=1.33s recent_step=1.52s eta=25:31:50 +2026-08-25 15:18:30 | INFO | train rank=0 | step=30840/100000 progress=30.8% loss=0.077484 lr=8.282929e-06 elapsed=11:22:59 avg_step=1.33s recent_step=1.41s eta=25:31:39 +2026-08-25 15:18:41 | INFO | train rank=0 | step=30850/100000 progress=30.9% loss=0.077667 lr=8.281681e-06 elapsed=11:23:11 avg_step=1.33s recent_step=1.14s eta=25:31:21 +2026-08-25 15:18:52 | INFO | train rank=0 | step=30860/100000 progress=30.9% loss=0.055618 lr=8.280434e-06 elapsed=11:23:22 avg_step=1.33s recent_step=1.13s eta=25:31:03 +2026-08-25 15:19:07 | INFO | train rank=0 | step=30870/100000 progress=30.9% loss=0.060303 lr=8.279186e-06 elapsed=11:23:36 avg_step=1.33s recent_step=1.43s eta=25:30:52 +2026-08-25 15:19:21 | INFO | train rank=0 | step=30880/100000 progress=30.9% loss=0.062103 lr=8.277937e-06 elapsed=11:23:51 avg_step=1.33s recent_step=1.43s eta=25:30:41 +2026-08-25 15:19:34 | INFO | train rank=0 | step=30890/100000 progress=30.9% loss=0.088257 lr=8.276689e-06 elapsed=11:24:04 avg_step=1.33s recent_step=1.36s eta=25:30:29 +2026-08-25 15:19:46 | INFO | train rank=0 | step=30900/100000 progress=30.9% loss=0.055420 lr=8.275440e-06 elapsed=11:24:16 avg_step=1.33s recent_step=1.15s eta=25:30:12 +2026-08-25 15:20:00 | INFO | train rank=0 | step=30910/100000 progress=30.9% loss=0.046417 lr=8.274190e-06 elapsed=11:24:30 avg_step=1.33s recent_step=1.39s eta=25:30:00 +2026-08-25 15:20:14 | INFO | train rank=0 | step=30920/100000 progress=30.9% loss=0.059219 lr=8.272940e-06 elapsed=11:24:44 avg_step=1.33s recent_step=1.44s eta=25:29:49 +2026-08-25 15:20:27 | INFO | train rank=0 | step=30930/100000 progress=30.9% loss=0.072174 lr=8.271690e-06 elapsed=11:24:57 avg_step=1.33s recent_step=1.23s eta=25:29:34 +2026-08-25 15:20:44 | INFO | train rank=0 | step=30940/100000 progress=30.9% loss=0.037720 lr=8.270440e-06 elapsed=11:25:14 avg_step=1.33s recent_step=1.77s eta=25:29:30 +2026-08-25 15:21:01 | INFO | train rank=0 | step=30950/100000 progress=30.9% loss=0.040482 lr=8.269189e-06 elapsed=11:25:31 avg_step=1.33s recent_step=1.63s eta=25:29:24 +2026-08-25 15:21:12 | INFO | train rank=0 | step=30960/100000 progress=31.0% loss=0.057465 lr=8.267937e-06 elapsed=11:25:42 avg_step=1.33s recent_step=1.13s eta=25:29:06 +2026-08-25 15:21:24 | INFO | train rank=0 | step=30970/100000 progress=31.0% loss=0.085388 lr=8.266686e-06 elapsed=11:25:54 avg_step=1.33s recent_step=1.20s eta=25:28:50 +2026-08-25 15:21:37 | INFO | train rank=0 | step=30980/100000 progress=31.0% loss=0.037155 lr=8.265434e-06 elapsed=11:26:07 avg_step=1.33s recent_step=1.34s eta=25:28:37 +2026-08-25 15:21:49 | INFO | train rank=0 | step=30990/100000 progress=31.0% loss=0.061615 lr=8.264181e-06 elapsed=11:26:19 avg_step=1.33s recent_step=1.18s eta=25:28:20 +2026-08-25 15:22:01 | INFO | train rank=0 | step=31000/100000 progress=31.0% loss=0.046738 lr=8.262929e-06 elapsed=11:26:31 avg_step=1.33s recent_step=1.15s eta=25:28:03 +2026-08-25 15:22:16 | INFO | train rank=0 | step=31010/100000 progress=31.0% loss=0.056564 lr=8.261676e-06 elapsed=11:26:46 avg_step=1.33s recent_step=1.53s eta=25:27:54 +2026-08-25 15:22:28 | INFO | train rank=0 | step=31020/100000 progress=31.0% loss=0.087097 lr=8.260422e-06 elapsed=11:26:58 avg_step=1.33s recent_step=1.23s eta=25:27:39 +2026-08-25 15:22:40 | INFO | train rank=0 | step=31030/100000 progress=31.0% loss=0.042587 lr=8.259169e-06 elapsed=11:27:10 avg_step=1.33s recent_step=1.17s eta=25:27:22 +2026-08-25 15:22:56 | INFO | train rank=0 | step=31040/100000 progress=31.0% loss=0.046585 lr=8.257915e-06 elapsed=11:27:26 avg_step=1.33s recent_step=1.64s eta=25:27:15 +2026-08-25 15:23:08 | INFO | train rank=0 | step=31050/100000 progress=31.1% loss=0.115967 lr=8.256660e-06 elapsed=11:27:38 avg_step=1.33s recent_step=1.19s eta=25:26:59 +2026-08-25 15:23:23 | INFO | train rank=0 | step=31060/100000 progress=31.1% loss=0.040627 lr=8.255405e-06 elapsed=11:27:53 avg_step=1.33s recent_step=1.48s eta=25:26:49 +2026-08-25 15:23:38 | INFO | train rank=0 | step=31070/100000 progress=31.1% loss=0.040970 lr=8.254150e-06 elapsed=11:28:08 avg_step=1.33s recent_step=1.51s eta=25:26:40 +2026-08-25 15:23:53 | INFO | train rank=0 | step=31080/100000 progress=31.1% loss=0.049408 lr=8.252895e-06 elapsed=11:28:23 avg_step=1.33s recent_step=1.44s eta=25:26:29 +2026-08-25 15:24:07 | INFO | train rank=0 | step=31090/100000 progress=31.1% loss=0.058777 lr=8.251639e-06 elapsed=11:28:37 avg_step=1.33s recent_step=1.40s eta=25:26:18 +2026-08-25 15:24:22 | INFO | train rank=0 | step=31100/100000 progress=31.1% loss=0.061768 lr=8.250382e-06 elapsed=11:28:52 avg_step=1.33s recent_step=1.52s eta=25:26:09 +2026-08-25 15:24:33 | INFO | train rank=0 | step=31110/100000 progress=31.1% loss=0.041016 lr=8.249126e-06 elapsed=11:29:03 avg_step=1.33s recent_step=1.15s eta=25:25:51 +2026-08-25 15:24:45 | INFO | train rank=0 | step=31120/100000 progress=31.1% loss=0.048218 lr=8.247869e-06 elapsed=11:29:15 avg_step=1.33s recent_step=1.12s eta=25:25:34 +2026-08-25 15:25:00 | INFO | train rank=0 | step=31130/100000 progress=31.1% loss=0.053467 lr=8.246612e-06 elapsed=11:29:30 avg_step=1.33s recent_step=1.53s eta=25:25:25 +2026-08-25 15:25:14 | INFO | train rank=0 | step=31140/100000 progress=31.1% loss=0.055756 lr=8.245354e-06 elapsed=11:29:44 avg_step=1.33s recent_step=1.44s eta=25:25:14 +2026-08-25 15:25:28 | INFO | train rank=0 | step=31150/100000 progress=31.1% loss=0.043045 lr=8.244096e-06 elapsed=11:29:58 avg_step=1.33s recent_step=1.39s eta=25:25:02 +2026-08-25 15:25:42 | INFO | train rank=0 | step=31160/100000 progress=31.2% loss=0.067657 lr=8.242837e-06 elapsed=11:30:11 avg_step=1.33s recent_step=1.33s eta=25:24:49 +2026-08-25 15:25:55 | INFO | train rank=0 | step=31170/100000 progress=31.2% loss=0.068542 lr=8.241579e-06 elapsed=11:30:25 avg_step=1.33s recent_step=1.39s eta=25:24:37 +2026-08-25 15:26:08 | INFO | train rank=0 | step=31180/100000 progress=31.2% loss=0.036140 lr=8.240320e-06 elapsed=11:30:37 avg_step=1.33s recent_step=1.20s eta=25:24:21 +2026-08-25 15:26:21 | INFO | train rank=0 | step=31190/100000 progress=31.2% loss=0.061157 lr=8.239060e-06 elapsed=11:30:50 avg_step=1.33s recent_step=1.31s eta=25:24:07 +2026-08-25 15:26:32 | INFO | train rank=0 | step=31200/100000 progress=31.2% loss=0.062622 lr=8.237800e-06 elapsed=11:31:01 avg_step=1.33s recent_step=1.09s eta=25:23:48 +2026-08-25 15:26:47 | INFO | train rank=0 | step=31210/100000 progress=31.2% loss=0.075958 lr=8.236540e-06 elapsed=11:31:17 avg_step=1.33s recent_step=1.58s eta=25:23:41 +2026-08-25 15:26:59 | INFO | train rank=0 | step=31220/100000 progress=31.2% loss=0.053604 lr=8.235280e-06 elapsed=11:31:29 avg_step=1.33s recent_step=1.14s eta=25:23:23 +2026-08-25 15:27:15 | INFO | train rank=0 | step=31230/100000 progress=31.2% loss=0.044540 lr=8.234019e-06 elapsed=11:31:45 avg_step=1.33s recent_step=1.63s eta=25:23:16 +2026-08-25 15:27:32 | INFO | train rank=0 | step=31240/100000 progress=31.2% loss=0.043518 lr=8.232758e-06 elapsed=11:32:02 avg_step=1.33s recent_step=1.70s eta=25:23:11 +2026-08-25 15:27:43 | INFO | train rank=0 | step=31250/100000 progress=31.2% loss=0.065811 lr=8.231496e-06 elapsed=11:32:13 avg_step=1.33s recent_step=1.10s eta=25:22:53 +2026-08-25 15:27:54 | INFO | train rank=0 | step=31260/100000 progress=31.3% loss=0.052094 lr=8.230234e-06 elapsed=11:32:24 avg_step=1.33s recent_step=1.15s eta=25:22:36 +2026-08-25 15:28:06 | INFO | train rank=0 | step=31270/100000 progress=31.3% loss=0.052307 lr=8.228972e-06 elapsed=11:32:36 avg_step=1.33s recent_step=1.14s eta=25:22:18 +2026-08-25 15:28:19 | INFO | train rank=0 | step=31280/100000 progress=31.3% loss=0.056381 lr=8.227709e-06 elapsed=11:32:49 avg_step=1.33s recent_step=1.30s eta=25:22:04 +2026-08-25 15:28:30 | INFO | train rank=0 | step=31290/100000 progress=31.3% loss=0.076721 lr=8.226446e-06 elapsed=11:33:00 avg_step=1.33s recent_step=1.16s eta=25:21:47 +2026-08-25 15:28:42 | INFO | train rank=0 | step=31300/100000 progress=31.3% loss=0.054886 lr=8.225183e-06 elapsed=11:33:12 avg_step=1.33s recent_step=1.12s eta=25:21:29 +2026-08-25 15:28:57 | INFO | train rank=0 | step=31310/100000 progress=31.3% loss=0.048676 lr=8.223919e-06 elapsed=11:33:27 avg_step=1.33s recent_step=1.50s eta=25:21:20 +2026-08-25 15:29:08 | INFO | train rank=0 | step=31320/100000 progress=31.3% loss=0.049767 lr=8.222655e-06 elapsed=11:33:38 avg_step=1.33s recent_step=1.16s eta=25:21:03 +2026-08-25 15:29:20 | INFO | train rank=0 | step=31330/100000 progress=31.3% loss=0.085907 lr=8.221391e-06 elapsed=11:33:50 avg_step=1.33s recent_step=1.17s eta=25:20:46 +2026-08-25 15:29:32 | INFO | train rank=0 | step=31340/100000 progress=31.3% loss=0.070679 lr=8.220126e-06 elapsed=11:34:02 avg_step=1.33s recent_step=1.17s eta=25:20:30 +2026-08-25 15:29:46 | INFO | train rank=0 | step=31350/100000 progress=31.4% loss=0.072845 lr=8.218861e-06 elapsed=11:34:16 avg_step=1.33s recent_step=1.41s eta=25:20:18 +2026-08-25 15:29:58 | INFO | train rank=0 | step=31360/100000 progress=31.4% loss=0.083008 lr=8.217596e-06 elapsed=11:34:28 avg_step=1.33s recent_step=1.22s eta=25:20:02 +2026-08-25 15:30:09 | INFO | train rank=0 | step=31370/100000 progress=31.4% loss=0.052948 lr=8.216330e-06 elapsed=11:34:39 avg_step=1.33s recent_step=1.14s eta=25:19:45 +2026-08-25 15:30:27 | INFO | train rank=0 | step=31380/100000 progress=31.4% loss=0.072418 lr=8.215064e-06 elapsed=11:34:57 avg_step=1.33s recent_step=1.74s eta=25:19:41 +2026-08-25 15:30:44 | INFO | train rank=0 | step=31390/100000 progress=31.4% loss=0.066833 lr=8.213797e-06 elapsed=11:35:14 avg_step=1.33s recent_step=1.71s eta=25:19:36 +2026-08-25 15:30:57 | INFO | train rank=0 | step=31400/100000 progress=31.4% loss=0.067383 lr=8.212530e-06 elapsed=11:35:26 avg_step=1.33s recent_step=1.27s eta=25:19:21 +2026-08-25 15:31:08 | INFO | train rank=0 | step=31410/100000 progress=31.4% loss=0.039703 lr=8.211263e-06 elapsed=11:35:38 avg_step=1.33s recent_step=1.12s eta=25:19:03 +2026-08-25 15:31:25 | INFO | train rank=0 | step=31420/100000 progress=31.4% loss=0.082733 lr=8.209996e-06 elapsed=11:35:54 avg_step=1.33s recent_step=1.68s eta=25:18:57 +2026-08-25 15:31:40 | INFO | train rank=0 | step=31430/100000 progress=31.4% loss=0.047485 lr=8.208728e-06 elapsed=11:36:10 avg_step=1.33s recent_step=1.57s eta=25:18:49 +2026-08-25 15:31:52 | INFO | train rank=0 | step=31440/100000 progress=31.4% loss=0.052170 lr=8.207460e-06 elapsed=11:36:22 avg_step=1.33s recent_step=1.14s eta=25:18:32 +2026-08-25 15:32:04 | INFO | train rank=0 | step=31450/100000 progress=31.4% loss=0.049728 lr=8.206191e-06 elapsed=11:36:34 avg_step=1.33s recent_step=1.21s eta=25:18:16 +2026-08-25 15:32:17 | INFO | train rank=0 | step=31460/100000 progress=31.5% loss=0.044159 lr=8.204922e-06 elapsed=11:36:47 avg_step=1.33s recent_step=1.37s eta=25:18:04 +2026-08-25 15:32:31 | INFO | train rank=0 | step=31470/100000 progress=31.5% loss=0.050110 lr=8.203653e-06 elapsed=11:37:01 avg_step=1.33s recent_step=1.38s eta=25:17:52 +2026-08-25 15:32:47 | INFO | train rank=0 | step=31480/100000 progress=31.5% loss=0.054718 lr=8.202383e-06 elapsed=11:37:17 avg_step=1.33s recent_step=1.61s eta=25:17:44 +2026-08-25 15:32:59 | INFO | train rank=0 | step=31490/100000 progress=31.5% loss=0.087524 lr=8.201113e-06 elapsed=11:37:29 avg_step=1.33s recent_step=1.16s eta=25:17:28 +2026-08-25 15:33:14 | INFO | train rank=0 | step=31500/100000 progress=31.5% loss=0.058075 lr=8.199843e-06 elapsed=11:37:44 avg_step=1.33s recent_step=1.52s eta=25:17:18 +2026-08-25 15:33:26 | INFO | train rank=0 | step=31510/100000 progress=31.5% loss=0.051163 lr=8.198572e-06 elapsed=11:37:56 avg_step=1.33s recent_step=1.16s eta=25:17:01 +2026-08-25 15:33:38 | INFO | train rank=0 | step=31520/100000 progress=31.5% loss=0.066193 lr=8.197301e-06 elapsed=11:38:08 avg_step=1.33s recent_step=1.27s eta=25:16:47 +2026-08-25 15:33:50 | INFO | train rank=0 | step=31530/100000 progress=31.5% loss=0.088074 lr=8.196030e-06 elapsed=11:38:20 avg_step=1.33s recent_step=1.19s eta=25:16:30 +2026-08-25 15:34:02 | INFO | train rank=0 | step=31540/100000 progress=31.5% loss=0.061264 lr=8.194758e-06 elapsed=11:38:32 avg_step=1.33s recent_step=1.15s eta=25:16:13 +2026-08-25 15:34:14 | INFO | train rank=0 | step=31550/100000 progress=31.6% loss=0.041718 lr=8.193486e-06 elapsed=11:38:44 avg_step=1.33s recent_step=1.23s eta=25:15:58 +2026-08-25 15:34:26 | INFO | train rank=0 | step=31560/100000 progress=31.6% loss=0.049667 lr=8.192213e-06 elapsed=11:38:56 avg_step=1.33s recent_step=1.18s eta=25:15:41 +2026-08-25 15:34:41 | INFO | train rank=0 | step=31570/100000 progress=31.6% loss=0.061829 lr=8.190940e-06 elapsed=11:39:11 avg_step=1.33s recent_step=1.46s eta=25:15:31 +2026-08-25 15:34:53 | INFO | train rank=0 | step=31580/100000 progress=31.6% loss=0.041229 lr=8.189667e-06 elapsed=11:39:23 avg_step=1.33s recent_step=1.29s eta=25:15:17 +2026-08-25 15:35:09 | INFO | train rank=0 | step=31590/100000 progress=31.6% loss=0.118347 lr=8.188394e-06 elapsed=11:39:39 avg_step=1.33s recent_step=1.55s eta=25:15:08 +2026-08-25 15:35:22 | INFO | train rank=0 | step=31600/100000 progress=31.6% loss=0.030083 lr=8.187120e-06 elapsed=11:39:52 avg_step=1.33s recent_step=1.29s eta=25:14:54 +2026-08-25 15:35:36 | INFO | train rank=0 | step=31610/100000 progress=31.6% loss=0.058655 lr=8.185846e-06 elapsed=11:40:06 avg_step=1.33s recent_step=1.44s eta=25:14:43 +2026-08-25 15:35:51 | INFO | train rank=0 | step=31620/100000 progress=31.6% loss=0.075317 lr=8.184571e-06 elapsed=11:40:21 avg_step=1.33s recent_step=1.45s eta=25:14:33 +2026-08-25 15:36:04 | INFO | train rank=0 | step=31630/100000 progress=31.6% loss=0.042740 lr=8.183296e-06 elapsed=11:40:34 avg_step=1.33s recent_step=1.30s eta=25:14:19 +2026-08-25 15:36:15 | INFO | train rank=0 | step=31640/100000 progress=31.6% loss=0.083496 lr=8.182021e-06 elapsed=11:40:45 avg_step=1.33s recent_step=1.14s eta=25:14:01 +2026-08-25 15:36:31 | INFO | train rank=0 | step=31650/100000 progress=31.6% loss=0.056892 lr=8.180745e-06 elapsed=11:41:01 avg_step=1.33s recent_step=1.62s eta=25:13:54 +2026-08-25 15:36:44 | INFO | train rank=0 | step=31660/100000 progress=31.7% loss=0.054672 lr=8.179470e-06 elapsed=11:41:13 avg_step=1.33s recent_step=1.22s eta=25:13:39 +2026-08-25 15:37:00 | INFO | train rank=0 | step=31670/100000 progress=31.7% loss=0.055557 lr=8.178193e-06 elapsed=11:41:30 avg_step=1.33s recent_step=1.67s eta=25:13:33 +2026-08-25 15:37:12 | INFO | train rank=0 | step=31680/100000 progress=31.7% loss=0.044586 lr=8.176917e-06 elapsed=11:41:42 avg_step=1.33s recent_step=1.16s eta=25:13:16 +2026-08-25 15:37:27 | INFO | train rank=0 | step=31690/100000 progress=31.7% loss=0.066528 lr=8.175640e-06 elapsed=11:41:57 avg_step=1.33s recent_step=1.52s eta=25:13:06 +2026-08-25 15:37:38 | INFO | train rank=0 | step=31700/100000 progress=31.7% loss=0.064896 lr=8.174362e-06 elapsed=11:42:08 avg_step=1.33s recent_step=1.11s eta=25:12:48 +2026-08-25 15:37:53 | INFO | train rank=0 | step=31710/100000 progress=31.7% loss=0.091187 lr=8.173085e-06 elapsed=11:42:23 avg_step=1.33s recent_step=1.48s eta=25:12:38 +2026-08-25 15:38:05 | INFO | train rank=0 | step=31720/100000 progress=31.7% loss=0.118469 lr=8.171807e-06 elapsed=11:42:35 avg_step=1.33s recent_step=1.19s eta=25:12:22 +2026-08-25 15:38:16 | INFO | train rank=0 | step=31730/100000 progress=31.7% loss=0.045761 lr=8.170528e-06 elapsed=11:42:46 avg_step=1.33s recent_step=1.16s eta=25:12:05 +2026-08-25 15:38:30 | INFO | train rank=0 | step=31740/100000 progress=31.7% loss=0.062180 lr=8.169250e-06 elapsed=11:43:00 avg_step=1.33s recent_step=1.34s eta=25:11:52 +2026-08-25 15:38:46 | INFO | train rank=0 | step=31750/100000 progress=31.8% loss=0.051743 lr=8.167971e-06 elapsed=11:43:15 avg_step=1.33s recent_step=1.58s eta=25:11:44 +2026-08-25 15:38:58 | INFO | train rank=0 | step=31760/100000 progress=31.8% loss=0.038033 lr=8.166691e-06 elapsed=11:43:28 avg_step=1.33s recent_step=1.21s eta=25:11:28 +2026-08-25 15:39:10 | INFO | train rank=0 | step=31770/100000 progress=31.8% loss=0.061050 lr=8.165411e-06 elapsed=11:43:40 avg_step=1.33s recent_step=1.25s eta=25:11:13 +2026-08-25 15:39:21 | INFO | train rank=0 | step=31780/100000 progress=31.8% loss=0.061707 lr=8.164131e-06 elapsed=11:43:51 avg_step=1.33s recent_step=1.14s eta=25:10:56 +2026-08-25 15:39:34 | INFO | train rank=0 | step=31790/100000 progress=31.8% loss=0.085602 lr=8.162851e-06 elapsed=11:44:04 avg_step=1.33s recent_step=1.27s eta=25:10:41 +2026-08-25 15:39:49 | INFO | train rank=0 | step=31800/100000 progress=31.8% loss=0.060562 lr=8.161570e-06 elapsed=11:44:19 avg_step=1.33s recent_step=1.51s eta=25:10:32 +2026-08-25 15:40:02 | INFO | train rank=0 | step=31810/100000 progress=31.8% loss=0.066559 lr=8.160289e-06 elapsed=11:44:32 avg_step=1.33s recent_step=1.26s eta=25:10:17 +2026-08-25 15:40:17 | INFO | train rank=0 | step=31820/100000 progress=31.8% loss=0.051544 lr=8.159007e-06 elapsed=11:44:46 avg_step=1.33s recent_step=1.46s eta=25:10:07 +2026-08-25 15:40:32 | INFO | train rank=0 | step=31830/100000 progress=31.8% loss=0.032669 lr=8.157726e-06 elapsed=11:45:01 avg_step=1.33s recent_step=1.50s eta=25:09:57 +2026-08-25 15:40:44 | INFO | train rank=0 | step=31840/100000 progress=31.8% loss=0.067642 lr=8.156443e-06 elapsed=11:45:14 avg_step=1.33s recent_step=1.24s eta=25:09:42 +2026-08-25 15:40:59 | INFO | train rank=0 | step=31850/100000 progress=31.9% loss=0.056961 lr=8.155161e-06 elapsed=11:45:28 avg_step=1.33s recent_step=1.46s eta=25:09:32 +2026-08-25 15:41:12 | INFO | train rank=0 | step=31860/100000 progress=31.9% loss=0.055557 lr=8.153878e-06 elapsed=11:45:41 avg_step=1.33s recent_step=1.30s eta=25:09:18 +2026-08-25 15:41:23 | INFO | train rank=0 | step=31870/100000 progress=31.9% loss=0.054749 lr=8.152595e-06 elapsed=11:45:53 avg_step=1.33s recent_step=1.16s eta=25:09:01 +2026-08-25 15:41:36 | INFO | train rank=0 | step=31880/100000 progress=31.9% loss=0.052795 lr=8.151311e-06 elapsed=11:46:06 avg_step=1.33s recent_step=1.34s eta=25:08:47 +2026-08-25 15:41:54 | INFO | train rank=0 | step=31890/100000 progress=31.9% loss=0.096619 lr=8.150027e-06 elapsed=11:46:24 avg_step=1.33s recent_step=1.71s eta=25:08:42 +2026-08-25 15:42:06 | INFO | train rank=0 | step=31900/100000 progress=31.9% loss=0.060287 lr=8.148743e-06 elapsed=11:46:35 avg_step=1.33s recent_step=1.19s eta=25:08:26 +2026-08-25 15:42:21 | INFO | train rank=0 | step=31910/100000 progress=31.9% loss=0.053223 lr=8.147459e-06 elapsed=11:46:51 avg_step=1.33s recent_step=1.56s eta=25:08:18 +2026-08-25 15:42:33 | INFO | train rank=0 | step=31920/100000 progress=31.9% loss=0.078339 lr=8.146174e-06 elapsed=11:47:03 avg_step=1.33s recent_step=1.17s eta=25:08:01 +2026-08-25 15:42:48 | INFO | train rank=0 | step=31930/100000 progress=31.9% loss=0.052124 lr=8.144888e-06 elapsed=11:47:18 avg_step=1.33s recent_step=1.49s eta=25:07:51 +2026-08-25 15:43:01 | INFO | train rank=0 | step=31940/100000 progress=31.9% loss=0.068680 lr=8.143603e-06 elapsed=11:47:31 avg_step=1.33s recent_step=1.35s eta=25:07:38 +2026-08-25 15:43:16 | INFO | train rank=0 | step=31950/100000 progress=31.9% loss=0.074829 lr=8.142317e-06 elapsed=11:47:46 avg_step=1.33s recent_step=1.51s eta=25:07:29 +2026-08-25 15:43:28 | INFO | train rank=0 | step=31960/100000 progress=32.0% loss=0.063354 lr=8.141030e-06 elapsed=11:47:58 avg_step=1.33s recent_step=1.15s eta=25:07:12 +2026-08-25 15:43:40 | INFO | train rank=0 | step=31970/100000 progress=32.0% loss=0.055298 lr=8.139744e-06 elapsed=11:48:10 avg_step=1.33s recent_step=1.24s eta=25:06:57 +2026-08-25 15:43:54 | INFO | train rank=0 | step=31980/100000 progress=32.0% loss=0.051086 lr=8.138457e-06 elapsed=11:48:24 avg_step=1.33s recent_step=1.35s eta=25:06:44 +2026-08-25 15:44:07 | INFO | train rank=0 | step=31990/100000 progress=32.0% loss=0.077240 lr=8.137169e-06 elapsed=11:48:37 avg_step=1.33s recent_step=1.35s eta=25:06:31 +2026-08-25 15:44:21 | INFO | train rank=0 | step=32000/100000 progress=32.0% loss=0.038055 lr=8.135882e-06 elapsed=11:48:51 avg_step=1.33s recent_step=1.37s eta=25:06:18 +2026-08-25 15:44:37 | INFO | train rank=0 | step=32010/100000 progress=32.0% loss=0.061844 lr=8.134594e-06 elapsed=11:49:06 avg_step=1.33s recent_step=1.57s eta=25:06:10 +2026-08-25 15:44:51 | INFO | train rank=0 | step=32020/100000 progress=32.0% loss=0.041656 lr=8.133305e-06 elapsed=11:49:20 avg_step=1.33s recent_step=1.40s eta=25:05:59 +2026-08-25 15:45:03 | INFO | train rank=0 | step=32030/100000 progress=32.0% loss=0.062943 lr=8.132017e-06 elapsed=11:49:33 avg_step=1.33s recent_step=1.27s eta=25:05:44 +2026-08-25 15:45:16 | INFO | train rank=0 | step=32040/100000 progress=32.0% loss=0.082703 lr=8.130728e-06 elapsed=11:49:46 avg_step=1.33s recent_step=1.31s eta=25:05:30 +2026-08-25 15:45:31 | INFO | train rank=0 | step=32050/100000 progress=32.0% loss=0.058228 lr=8.129438e-06 elapsed=11:50:01 avg_step=1.33s recent_step=1.47s eta=25:05:20 +2026-08-25 15:45:43 | INFO | train rank=0 | step=32060/100000 progress=32.1% loss=0.030891 lr=8.128149e-06 elapsed=11:50:13 avg_step=1.33s recent_step=1.23s eta=25:05:05 +2026-08-25 15:45:55 | INFO | train rank=0 | step=32070/100000 progress=32.1% loss=0.065552 lr=8.126858e-06 elapsed=11:50:25 avg_step=1.33s recent_step=1.13s eta=25:04:47 +2026-08-25 15:46:06 | INFO | train rank=0 | step=32080/100000 progress=32.1% loss=0.047325 lr=8.125568e-06 elapsed=11:50:36 avg_step=1.33s recent_step=1.15s eta=25:04:30 +2026-08-25 15:46:20 | INFO | train rank=0 | step=32090/100000 progress=32.1% loss=0.060944 lr=8.124277e-06 elapsed=11:50:50 avg_step=1.33s recent_step=1.35s eta=25:04:17 +2026-08-25 15:46:36 | INFO | train rank=0 | step=32100/100000 progress=32.1% loss=0.046051 lr=8.122986e-06 elapsed=11:51:06 avg_step=1.33s recent_step=1.60s eta=25:04:09 +2026-08-25 15:46:48 | INFO | train rank=0 | step=32110/100000 progress=32.1% loss=0.037903 lr=8.121695e-06 elapsed=11:51:18 avg_step=1.33s recent_step=1.21s eta=25:03:54 +2026-08-25 15:47:00 | INFO | train rank=0 | step=32120/100000 progress=32.1% loss=0.092896 lr=8.120403e-06 elapsed=11:51:30 avg_step=1.33s recent_step=1.20s eta=25:03:38 +2026-08-25 15:47:11 | INFO | train rank=0 | step=32130/100000 progress=32.1% loss=0.075684 lr=8.119111e-06 elapsed=11:51:41 avg_step=1.33s recent_step=1.15s eta=25:03:21 +2026-08-25 15:47:29 | INFO | train rank=0 | step=32140/100000 progress=32.1% loss=0.073242 lr=8.117818e-06 elapsed=11:51:59 avg_step=1.33s recent_step=1.78s eta=25:03:17 +2026-08-25 15:47:45 | INFO | train rank=0 | step=32150/100000 progress=32.1% loss=0.081757 lr=8.116526e-06 elapsed=11:52:15 avg_step=1.33s recent_step=1.55s eta=25:03:08 +2026-08-25 15:47:57 | INFO | train rank=0 | step=32160/100000 progress=32.2% loss=0.056473 lr=8.115232e-06 elapsed=11:52:27 avg_step=1.33s recent_step=1.28s eta=25:02:54 +2026-08-25 15:48:12 | INFO | train rank=0 | step=32170/100000 progress=32.2% loss=0.065948 lr=8.113939e-06 elapsed=11:52:42 avg_step=1.33s recent_step=1.50s eta=25:02:44 +2026-08-25 15:48:28 | INFO | train rank=0 | step=32180/100000 progress=32.2% loss=0.052124 lr=8.112645e-06 elapsed=11:52:58 avg_step=1.33s recent_step=1.56s eta=25:02:36 +2026-08-25 15:48:39 | INFO | train rank=0 | step=32190/100000 progress=32.2% loss=0.043579 lr=8.111351e-06 elapsed=11:53:09 avg_step=1.33s recent_step=1.10s eta=25:02:18 +2026-08-25 15:48:51 | INFO | train rank=0 | step=32200/100000 progress=32.2% loss=0.052673 lr=8.110056e-06 elapsed=11:53:20 avg_step=1.33s recent_step=1.15s eta=25:02:01 +2026-08-25 15:49:02 | INFO | train rank=0 | step=32210/100000 progress=32.2% loss=0.119080 lr=8.108762e-06 elapsed=11:53:32 avg_step=1.33s recent_step=1.16s eta=25:01:44 +2026-08-25 15:49:17 | INFO | train rank=0 | step=32220/100000 progress=32.2% loss=0.044846 lr=8.107466e-06 elapsed=11:53:47 avg_step=1.33s recent_step=1.47s eta=25:01:34 +2026-08-25 15:49:29 | INFO | train rank=0 | step=32230/100000 progress=32.2% loss=0.051003 lr=8.106171e-06 elapsed=11:53:59 avg_step=1.33s recent_step=1.22s eta=25:01:18 +2026-08-25 15:49:40 | INFO | train rank=0 | step=32240/100000 progress=32.2% loss=0.057312 lr=8.104875e-06 elapsed=11:54:10 avg_step=1.33s recent_step=1.11s eta=25:01:00 +2026-08-25 15:49:52 | INFO | train rank=0 | step=32250/100000 progress=32.2% loss=0.049835 lr=8.103579e-06 elapsed=11:54:22 avg_step=1.33s recent_step=1.15s eta=25:00:43 +2026-08-25 15:50:08 | INFO | train rank=0 | step=32260/100000 progress=32.3% loss=0.030571 lr=8.102282e-06 elapsed=11:54:38 avg_step=1.33s recent_step=1.65s eta=25:00:36 +2026-08-25 15:50:20 | INFO | train rank=0 | step=32270/100000 progress=32.3% loss=0.043411 lr=8.100985e-06 elapsed=11:54:50 avg_step=1.33s recent_step=1.19s eta=25:00:20 +2026-08-25 15:50:34 | INFO | train rank=0 | step=32280/100000 progress=32.3% loss=0.063751 lr=8.099688e-06 elapsed=11:55:04 avg_step=1.33s recent_step=1.41s eta=25:00:09 +2026-08-25 15:50:50 | INFO | train rank=0 | step=32290/100000 progress=32.3% loss=0.065140 lr=8.098391e-06 elapsed=11:55:20 avg_step=1.33s recent_step=1.62s eta=25:00:01 +2026-08-25 15:51:04 | INFO | train rank=0 | step=32300/100000 progress=32.3% loss=0.074219 lr=8.097093e-06 elapsed=11:55:34 avg_step=1.33s recent_step=1.34s eta=24:59:48 +2026-08-25 15:51:15 | INFO | train rank=0 | step=32310/100000 progress=32.3% loss=0.042828 lr=8.095794e-06 elapsed=11:55:45 avg_step=1.33s recent_step=1.17s eta=24:59:32 +2026-08-25 15:51:27 | INFO | train rank=0 | step=32320/100000 progress=32.3% loss=0.045471 lr=8.094496e-06 elapsed=11:55:57 avg_step=1.33s recent_step=1.17s eta=24:59:15 +2026-08-25 15:51:39 | INFO | train rank=0 | step=32330/100000 progress=32.3% loss=0.039749 lr=8.093197e-06 elapsed=11:56:09 avg_step=1.33s recent_step=1.19s eta=24:58:59 +2026-08-25 15:51:50 | INFO | train rank=0 | step=32340/100000 progress=32.3% loss=0.042358 lr=8.091898e-06 elapsed=11:56:20 avg_step=1.33s recent_step=1.10s eta=24:58:41 +2026-08-25 15:52:10 | INFO | train rank=0 | step=32350/100000 progress=32.4% loss=0.050873 lr=8.090598e-06 elapsed=11:56:39 avg_step=1.33s recent_step=1.94s eta=24:58:40 +2026-08-25 15:52:21 | INFO | train rank=0 | step=32360/100000 progress=32.4% loss=0.081268 lr=8.089298e-06 elapsed=11:56:51 avg_step=1.33s recent_step=1.16s eta=24:58:24 +2026-08-25 15:52:34 | INFO | train rank=0 | step=32370/100000 progress=32.4% loss=0.083984 lr=8.087998e-06 elapsed=11:57:04 avg_step=1.33s recent_step=1.26s eta=24:58:09 +2026-08-25 15:52:47 | INFO | train rank=0 | step=32380/100000 progress=32.4% loss=0.049042 lr=8.086697e-06 elapsed=11:57:17 avg_step=1.33s recent_step=1.37s eta=24:57:56 +2026-08-25 15:52:59 | INFO | train rank=0 | step=32390/100000 progress=32.4% loss=0.048798 lr=8.085396e-06 elapsed=11:57:29 avg_step=1.33s recent_step=1.13s eta=24:57:39 +2026-08-25 15:53:11 | INFO | train rank=0 | step=32400/100000 progress=32.4% loss=0.063782 lr=8.084095e-06 elapsed=11:57:41 avg_step=1.33s recent_step=1.21s eta=24:57:23 +2026-08-25 15:53:26 | INFO | train rank=0 | step=32410/100000 progress=32.4% loss=0.059570 lr=8.082793e-06 elapsed=11:57:56 avg_step=1.33s recent_step=1.50s eta=24:57:13 +2026-08-25 15:53:38 | INFO | train rank=0 | step=32420/100000 progress=32.4% loss=0.067200 lr=8.081491e-06 elapsed=11:58:08 avg_step=1.33s recent_step=1.19s eta=24:56:57 +2026-08-25 15:53:52 | INFO | train rank=0 | step=32430/100000 progress=32.4% loss=0.034119 lr=8.080189e-06 elapsed=11:58:22 avg_step=1.33s recent_step=1.42s eta=24:56:46 +2026-08-25 15:54:03 | INFO | train rank=0 | step=32440/100000 progress=32.4% loss=0.054382 lr=8.078886e-06 elapsed=11:58:33 avg_step=1.33s recent_step=1.13s eta=24:56:28 +2026-08-25 15:54:18 | INFO | train rank=0 | step=32450/100000 progress=32.5% loss=0.056946 lr=8.077583e-06 elapsed=11:58:47 avg_step=1.33s recent_step=1.45s eta=24:56:17 +2026-08-25 15:54:36 | INFO | train rank=0 | step=32460/100000 progress=32.5% loss=0.026932 lr=8.076280e-06 elapsed=11:59:06 avg_step=1.33s recent_step=1.81s eta=24:56:14 +2026-08-25 15:54:47 | INFO | train rank=0 | step=32470/100000 progress=32.5% loss=0.078644 lr=8.074977e-06 elapsed=11:59:17 avg_step=1.33s recent_step=1.15s eta=24:55:57 +2026-08-25 15:54:59 | INFO | train rank=0 | step=32480/100000 progress=32.5% loss=0.053741 lr=8.073673e-06 elapsed=11:59:29 avg_step=1.33s recent_step=1.17s eta=24:55:41 +2026-08-25 15:55:13 | INFO | train rank=0 | step=32490/100000 progress=32.5% loss=0.035233 lr=8.072368e-06 elapsed=11:59:43 avg_step=1.33s recent_step=1.44s eta=24:55:30 +2026-08-25 15:55:26 | INFO | train rank=0 | step=32500/100000 progress=32.5% loss=0.044952 lr=8.071064e-06 elapsed=11:59:56 avg_step=1.33s recent_step=1.32s eta=24:55:16 +2026-08-25 15:55:40 | INFO | train rank=0 | step=32510/100000 progress=32.5% loss=0.044281 lr=8.069759e-06 elapsed=12:00:10 avg_step=1.33s recent_step=1.40s eta=24:55:04 +2026-08-25 15:55:57 | INFO | train rank=0 | step=32520/100000 progress=32.5% loss=0.068512 lr=8.068453e-06 elapsed=12:00:27 avg_step=1.33s recent_step=1.64s eta=24:54:57 +2026-08-25 15:56:09 | INFO | train rank=0 | step=32530/100000 progress=32.5% loss=0.078430 lr=8.067148e-06 elapsed=12:00:38 avg_step=1.33s recent_step=1.17s eta=24:54:41 +2026-08-25 15:56:20 | INFO | train rank=0 | step=32540/100000 progress=32.5% loss=0.058990 lr=8.065842e-06 elapsed=12:00:50 avg_step=1.33s recent_step=1.16s eta=24:54:24 +2026-08-25 15:56:33 | INFO | train rank=0 | step=32550/100000 progress=32.6% loss=0.052086 lr=8.064535e-06 elapsed=12:01:03 avg_step=1.33s recent_step=1.33s eta=24:54:11 +2026-08-25 15:56:47 | INFO | train rank=0 | step=32560/100000 progress=32.6% loss=0.056671 lr=8.063229e-06 elapsed=12:01:17 avg_step=1.33s recent_step=1.39s eta=24:53:59 +2026-08-25 15:57:01 | INFO | train rank=0 | step=32570/100000 progress=32.6% loss=0.058868 lr=8.061922e-06 elapsed=12:01:30 avg_step=1.33s recent_step=1.32s eta=24:53:45 +2026-08-25 15:57:12 | INFO | train rank=0 | step=32580/100000 progress=32.6% loss=0.072922 lr=8.060614e-06 elapsed=12:01:42 avg_step=1.33s recent_step=1.13s eta=24:53:28 +2026-08-25 15:57:24 | INFO | train rank=0 | step=32590/100000 progress=32.6% loss=0.077789 lr=8.059307e-06 elapsed=12:01:54 avg_step=1.33s recent_step=1.17s eta=24:53:11 +2026-08-25 15:57:38 | INFO | train rank=0 | step=32600/100000 progress=32.6% loss=0.105255 lr=8.057999e-06 elapsed=12:02:08 avg_step=1.33s recent_step=1.44s eta=24:53:00 +2026-08-25 15:57:51 | INFO | train rank=0 | step=32610/100000 progress=32.6% loss=0.049728 lr=8.056690e-06 elapsed=12:02:21 avg_step=1.33s recent_step=1.28s eta=24:52:46 +2026-08-25 15:58:04 | INFO | train rank=0 | step=32620/100000 progress=32.6% loss=0.050247 lr=8.055382e-06 elapsed=12:02:34 avg_step=1.33s recent_step=1.28s eta=24:52:32 +2026-08-25 15:58:18 | INFO | train rank=0 | step=32630/100000 progress=32.6% loss=0.044220 lr=8.054073e-06 elapsed=12:02:48 avg_step=1.33s recent_step=1.45s eta=24:52:21 +2026-08-25 15:58:36 | INFO | train rank=0 | step=32640/100000 progress=32.6% loss=0.077271 lr=8.052763e-06 elapsed=12:03:06 avg_step=1.33s recent_step=1.76s eta=24:52:17 +2026-08-25 15:58:47 | INFO | train rank=0 | step=32650/100000 progress=32.6% loss=0.063599 lr=8.051454e-06 elapsed=12:03:17 avg_step=1.33s recent_step=1.12s eta=24:51:59 +2026-08-25 15:59:02 | INFO | train rank=0 | step=32660/100000 progress=32.7% loss=0.070953 lr=8.050144e-06 elapsed=12:03:32 avg_step=1.33s recent_step=1.47s eta=24:51:49 +2026-08-25 15:59:13 | INFO | train rank=0 | step=32670/100000 progress=32.7% loss=0.060211 lr=8.048833e-06 elapsed=12:03:43 avg_step=1.33s recent_step=1.16s eta=24:51:32 +2026-08-25 15:59:31 | INFO | train rank=0 | step=32680/100000 progress=32.7% loss=0.067963 lr=8.047523e-06 elapsed=12:04:01 avg_step=1.33s recent_step=1.79s eta=24:51:28 +2026-08-25 15:59:43 | INFO | train rank=0 | step=32690/100000 progress=32.7% loss=0.061859 lr=8.046212e-06 elapsed=12:04:13 avg_step=1.33s recent_step=1.15s eta=24:51:11 +2026-08-25 15:59:55 | INFO | train rank=0 | step=32700/100000 progress=32.7% loss=0.087769 lr=8.044900e-06 elapsed=12:04:25 avg_step=1.33s recent_step=1.27s eta=24:50:57 +2026-08-25 16:00:10 | INFO | train rank=0 | step=32710/100000 progress=32.7% loss=0.035683 lr=8.043589e-06 elapsed=12:04:40 avg_step=1.33s recent_step=1.43s eta=24:50:46 +2026-08-25 16:00:21 | INFO | train rank=0 | step=32720/100000 progress=32.7% loss=0.060669 lr=8.042277e-06 elapsed=12:04:51 avg_step=1.33s recent_step=1.17s eta=24:50:29 +2026-08-25 16:00:36 | INFO | train rank=0 | step=32730/100000 progress=32.7% loss=0.077057 lr=8.040964e-06 elapsed=12:05:06 avg_step=1.33s recent_step=1.50s eta=24:50:19 +2026-08-25 16:00:48 | INFO | train rank=0 | step=32740/100000 progress=32.7% loss=0.050110 lr=8.039651e-06 elapsed=12:05:18 avg_step=1.33s recent_step=1.15s eta=24:50:02 +2026-08-25 16:01:01 | INFO | train rank=0 | step=32750/100000 progress=32.8% loss=0.053146 lr=8.038338e-06 elapsed=12:05:31 avg_step=1.33s recent_step=1.32s eta=24:49:48 +2026-08-25 16:01:13 | INFO | train rank=0 | step=32760/100000 progress=32.8% loss=0.114380 lr=8.037025e-06 elapsed=12:05:42 avg_step=1.33s recent_step=1.15s eta=24:49:32 +2026-08-25 16:01:28 | INFO | train rank=0 | step=32770/100000 progress=32.8% loss=0.071060 lr=8.035711e-06 elapsed=12:05:58 avg_step=1.33s recent_step=1.53s eta=24:49:22 +2026-08-25 16:01:40 | INFO | train rank=0 | step=32780/100000 progress=32.8% loss=0.053574 lr=8.034397e-06 elapsed=12:06:10 avg_step=1.33s recent_step=1.18s eta=24:49:06 +2026-08-25 16:01:52 | INFO | train rank=0 | step=32790/100000 progress=32.8% loss=0.045486 lr=8.033083e-06 elapsed=12:06:22 avg_step=1.33s recent_step=1.25s eta=24:48:51 +2026-08-25 16:02:04 | INFO | train rank=0 | step=32800/100000 progress=32.8% loss=0.069824 lr=8.031768e-06 elapsed=12:06:34 avg_step=1.33s recent_step=1.15s eta=24:48:34 +2026-08-25 16:02:18 | INFO | train rank=0 | step=32810/100000 progress=32.8% loss=0.054138 lr=8.030453e-06 elapsed=12:06:47 avg_step=1.33s recent_step=1.39s eta=24:48:22 +2026-08-25 16:02:30 | INFO | train rank=0 | step=32820/100000 progress=32.8% loss=0.087372 lr=8.029138e-06 elapsed=12:07:00 avg_step=1.33s recent_step=1.22s eta=24:48:07 +2026-08-25 16:02:41 | INFO | train rank=0 | step=32830/100000 progress=32.8% loss=0.054398 lr=8.027822e-06 elapsed=12:07:11 avg_step=1.33s recent_step=1.16s eta=24:47:50 +2026-08-25 16:02:56 | INFO | train rank=0 | step=32840/100000 progress=32.8% loss=0.051239 lr=8.026507e-06 elapsed=12:07:26 avg_step=1.33s recent_step=1.43s eta=24:47:39 +2026-08-25 16:03:09 | INFO | train rank=0 | step=32850/100000 progress=32.9% loss=0.034454 lr=8.025190e-06 elapsed=12:07:39 avg_step=1.33s recent_step=1.32s eta=24:47:25 +2026-08-25 16:03:23 | INFO | train rank=0 | step=32860/100000 progress=32.9% loss=0.067413 lr=8.023874e-06 elapsed=12:07:53 avg_step=1.33s recent_step=1.38s eta=24:47:13 +2026-08-25 16:03:34 | INFO | train rank=0 | step=32870/100000 progress=32.9% loss=0.067688 lr=8.022557e-06 elapsed=12:08:04 avg_step=1.33s recent_step=1.15s eta=24:46:56 +2026-08-25 16:03:49 | INFO | train rank=0 | step=32880/100000 progress=32.9% loss=0.077026 lr=8.021239e-06 elapsed=12:08:18 avg_step=1.33s recent_step=1.43s eta=24:46:45 +2026-08-25 16:04:01 | INFO | train rank=0 | step=32890/100000 progress=32.9% loss=0.045746 lr=8.019922e-06 elapsed=12:08:31 avg_step=1.33s recent_step=1.23s eta=24:46:29 +2026-08-25 16:04:17 | INFO | train rank=0 | step=32900/100000 progress=32.9% loss=0.043060 lr=8.018604e-06 elapsed=12:08:47 avg_step=1.33s recent_step=1.59s eta=24:46:21 +2026-08-25 16:04:28 | INFO | train rank=0 | step=32910/100000 progress=32.9% loss=0.062866 lr=8.017285e-06 elapsed=12:08:58 avg_step=1.33s recent_step=1.17s eta=24:46:05 +2026-08-25 16:04:47 | INFO | train rank=0 | step=32920/100000 progress=32.9% loss=0.064728 lr=8.015967e-06 elapsed=12:09:17 avg_step=1.33s recent_step=1.84s eta=24:46:02 +2026-08-25 16:04:59 | INFO | train rank=0 | step=32930/100000 progress=32.9% loss=0.053513 lr=8.014648e-06 elapsed=12:09:29 avg_step=1.33s recent_step=1.21s eta=24:45:46 +2026-08-25 16:05:11 | INFO | train rank=0 | step=32940/100000 progress=32.9% loss=0.045609 lr=8.013328e-06 elapsed=12:09:41 avg_step=1.33s recent_step=1.20s eta=24:45:30 +2026-08-25 16:05:25 | INFO | train rank=0 | step=32950/100000 progress=33.0% loss=0.110046 lr=8.012009e-06 elapsed=12:09:55 avg_step=1.33s recent_step=1.40s eta=24:45:18 +2026-08-25 16:05:39 | INFO | train rank=0 | step=32960/100000 progress=33.0% loss=0.108093 lr=8.010689e-06 elapsed=12:10:09 avg_step=1.33s recent_step=1.42s eta=24:45:07 +2026-08-25 16:05:54 | INFO | train rank=0 | step=32970/100000 progress=33.0% loss=0.054886 lr=8.009369e-06 elapsed=12:10:24 avg_step=1.33s recent_step=1.47s eta=24:44:57 +2026-08-25 16:06:06 | INFO | train rank=0 | step=32980/100000 progress=33.0% loss=0.054764 lr=8.008048e-06 elapsed=12:10:36 avg_step=1.33s recent_step=1.23s eta=24:44:41 +2026-08-25 16:06:18 | INFO | train rank=0 | step=32990/100000 progress=33.0% loss=0.079712 lr=8.006727e-06 elapsed=12:10:48 avg_step=1.33s recent_step=1.19s eta=24:44:25 +2026-08-25 16:06:32 | INFO | train rank=0 | step=33000/100000 progress=33.0% loss=0.060974 lr=8.005406e-06 elapsed=12:11:02 avg_step=1.33s recent_step=1.46s eta=24:44:14 +2026-08-25 16:06:46 | INFO | train rank=0 | step=33010/100000 progress=33.0% loss=0.035362 lr=8.004084e-06 elapsed=12:11:16 avg_step=1.33s recent_step=1.40s eta=24:44:03 +2026-08-25 16:06:58 | INFO | train rank=0 | step=33020/100000 progress=33.0% loss=0.044891 lr=8.002762e-06 elapsed=12:11:28 avg_step=1.33s recent_step=1.19s eta=24:43:46 +2026-08-25 16:07:10 | INFO | train rank=0 | step=33030/100000 progress=33.0% loss=0.053024 lr=8.001440e-06 elapsed=12:11:40 avg_step=1.33s recent_step=1.14s eta=24:43:29 +2026-08-25 16:07:22 | INFO | train rank=0 | step=33040/100000 progress=33.0% loss=0.053680 lr=8.000117e-06 elapsed=12:11:52 avg_step=1.33s recent_step=1.25s eta=24:43:15 +2026-08-25 16:07:38 | INFO | train rank=0 | step=33050/100000 progress=33.1% loss=0.067810 lr=7.998794e-06 elapsed=12:12:08 avg_step=1.33s recent_step=1.55s eta=24:43:06 +2026-08-25 16:07:49 | INFO | train rank=0 | step=33060/100000 progress=33.1% loss=0.078064 lr=7.997471e-06 elapsed=12:12:19 avg_step=1.33s recent_step=1.17s eta=24:42:49 +2026-08-25 16:08:03 | INFO | train rank=0 | step=33070/100000 progress=33.1% loss=0.069000 lr=7.996148e-06 elapsed=12:12:33 avg_step=1.33s recent_step=1.35s eta=24:42:36 +2026-08-25 16:08:16 | INFO | train rank=0 | step=33080/100000 progress=33.1% loss=0.058167 lr=7.994824e-06 elapsed=12:12:46 avg_step=1.33s recent_step=1.29s eta=24:42:22 +2026-08-25 16:08:28 | INFO | train rank=0 | step=33090/100000 progress=33.1% loss=0.043243 lr=7.993500e-06 elapsed=12:12:58 avg_step=1.33s recent_step=1.24s eta=24:42:07 +2026-08-25 16:08:40 | INFO | train rank=0 | step=33100/100000 progress=33.1% loss=0.048462 lr=7.992175e-06 elapsed=12:13:10 avg_step=1.33s recent_step=1.19s eta=24:41:51 +2026-08-25 16:08:54 | INFO | train rank=0 | step=33110/100000 progress=33.1% loss=0.041519 lr=7.990850e-06 elapsed=12:13:24 avg_step=1.33s recent_step=1.42s eta=24:41:39 +2026-08-25 16:09:08 | INFO | train rank=0 | step=33120/100000 progress=33.1% loss=0.047546 lr=7.989525e-06 elapsed=12:13:38 avg_step=1.33s recent_step=1.36s eta=24:41:27 +2026-08-25 16:09:23 | INFO | train rank=0 | step=33130/100000 progress=33.1% loss=0.043457 lr=7.988199e-06 elapsed=12:13:53 avg_step=1.33s recent_step=1.47s eta=24:41:16 +2026-08-25 16:09:41 | INFO | train rank=0 | step=33140/100000 progress=33.1% loss=0.075439 lr=7.986874e-06 elapsed=12:14:11 avg_step=1.33s recent_step=1.83s eta=24:41:13 +2026-08-25 16:09:53 | INFO | train rank=0 | step=33150/100000 progress=33.1% loss=0.066864 lr=7.985547e-06 elapsed=12:14:23 avg_step=1.33s recent_step=1.18s eta=24:40:57 +2026-08-25 16:10:05 | INFO | train rank=0 | step=33160/100000 progress=33.2% loss=0.036743 lr=7.984221e-06 elapsed=12:14:35 avg_step=1.33s recent_step=1.25s eta=24:40:42 +2026-08-25 16:10:18 | INFO | train rank=0 | step=33170/100000 progress=33.2% loss=0.043762 lr=7.982894e-06 elapsed=12:14:48 avg_step=1.33s recent_step=1.25s eta=24:40:27 +2026-08-25 16:10:29 | INFO | train rank=0 | step=33180/100000 progress=33.2% loss=0.078949 lr=7.981567e-06 elapsed=12:14:59 avg_step=1.33s recent_step=1.11s eta=24:40:09 +2026-08-25 16:10:43 | INFO | train rank=0 | step=33190/100000 progress=33.2% loss=0.065491 lr=7.980239e-06 elapsed=12:15:13 avg_step=1.33s recent_step=1.42s eta=24:39:58 +2026-08-25 16:10:55 | INFO | train rank=0 | step=33200/100000 progress=33.2% loss=0.068741 lr=7.978912e-06 elapsed=12:15:25 avg_step=1.33s recent_step=1.22s eta=24:39:42 +2026-08-25 16:11:08 | INFO | train rank=0 | step=33210/100000 progress=33.2% loss=0.056854 lr=7.977583e-06 elapsed=12:15:38 avg_step=1.33s recent_step=1.26s eta=24:39:28 +2026-08-25 16:11:22 | INFO | train rank=0 | step=33220/100000 progress=33.2% loss=0.039032 lr=7.976255e-06 elapsed=12:15:52 avg_step=1.33s recent_step=1.41s eta=24:39:16 +2026-08-25 16:11:36 | INFO | train rank=0 | step=33230/100000 progress=33.2% loss=0.049423 lr=7.974926e-06 elapsed=12:16:06 avg_step=1.33s recent_step=1.37s eta=24:39:04 +2026-08-25 16:11:54 | INFO | train rank=0 | step=33240/100000 progress=33.2% loss=0.055405 lr=7.973597e-06 elapsed=12:16:23 avg_step=1.33s recent_step=1.79s eta=24:39:00 +2026-08-25 16:12:05 | INFO | train rank=0 | step=33250/100000 progress=33.2% loss=0.048126 lr=7.972268e-06 elapsed=12:16:35 avg_step=1.33s recent_step=1.15s eta=24:38:43 +2026-08-25 16:12:16 | INFO | train rank=0 | step=33260/100000 progress=33.3% loss=0.070374 lr=7.970938e-06 elapsed=12:16:46 avg_step=1.33s recent_step=1.11s eta=24:38:25 +2026-08-25 16:12:27 | INFO | train rank=0 | step=33270/100000 progress=33.3% loss=0.056946 lr=7.969608e-06 elapsed=12:16:57 avg_step=1.33s recent_step=1.13s eta=24:38:08 +2026-08-25 16:12:45 | INFO | train rank=0 | step=33280/100000 progress=33.3% loss=0.061951 lr=7.968277e-06 elapsed=12:17:15 avg_step=1.33s recent_step=1.74s eta=24:38:03 +2026-08-25 16:12:56 | INFO | train rank=0 | step=33290/100000 progress=33.3% loss=0.079590 lr=7.966947e-06 elapsed=12:17:26 avg_step=1.33s recent_step=1.10s eta=24:37:45 +2026-08-25 16:13:11 | INFO | train rank=0 | step=33300/100000 progress=33.3% loss=0.078369 lr=7.965615e-06 elapsed=12:17:41 avg_step=1.33s recent_step=1.55s eta=24:37:36 +2026-08-25 16:13:22 | INFO | train rank=0 | step=33310/100000 progress=33.3% loss=0.067490 lr=7.964284e-06 elapsed=12:17:52 avg_step=1.33s recent_step=1.07s eta=24:37:17 +2026-08-25 16:13:38 | INFO | train rank=0 | step=33320/100000 progress=33.3% loss=0.043999 lr=7.962952e-06 elapsed=12:18:08 avg_step=1.33s recent_step=1.59s eta=24:37:09 +2026-08-25 16:13:49 | INFO | train rank=0 | step=33330/100000 progress=33.3% loss=0.049500 lr=7.961620e-06 elapsed=12:18:19 avg_step=1.33s recent_step=1.15s eta=24:36:52 +2026-08-25 16:14:03 | INFO | train rank=0 | step=33340/100000 progress=33.3% loss=0.048996 lr=7.960288e-06 elapsed=12:18:33 avg_step=1.33s recent_step=1.37s eta=24:36:40 +2026-08-25 16:14:21 | INFO | train rank=0 | step=33350/100000 progress=33.4% loss=0.102051 lr=7.958955e-06 elapsed=12:18:51 avg_step=1.33s recent_step=1.81s eta=24:36:36 +2026-08-25 16:14:32 | INFO | train rank=0 | step=33360/100000 progress=33.4% loss=0.076874 lr=7.957622e-06 elapsed=12:19:02 avg_step=1.33s recent_step=1.10s eta=24:36:18 +2026-08-25 16:14:44 | INFO | train rank=0 | step=33370/100000 progress=33.4% loss=0.074280 lr=7.956289e-06 elapsed=12:19:14 avg_step=1.33s recent_step=1.14s eta=24:36:01 +2026-08-25 16:14:56 | INFO | train rank=0 | step=33380/100000 progress=33.4% loss=0.083755 lr=7.954955e-06 elapsed=12:19:26 avg_step=1.33s recent_step=1.23s eta=24:35:46 +2026-08-25 16:15:07 | INFO | train rank=0 | step=33390/100000 progress=33.4% loss=0.041245 lr=7.953621e-06 elapsed=12:19:37 avg_step=1.33s recent_step=1.12s eta=24:35:29 +2026-08-25 16:15:19 | INFO | train rank=0 | step=33400/100000 progress=33.4% loss=0.052094 lr=7.952287e-06 elapsed=12:19:49 avg_step=1.33s recent_step=1.17s eta=24:35:12 +2026-08-25 16:15:32 | INFO | train rank=0 | step=33410/100000 progress=33.4% loss=0.062714 lr=7.950952e-06 elapsed=12:20:02 avg_step=1.33s recent_step=1.30s eta=24:34:58 +2026-08-25 16:15:47 | INFO | train rank=0 | step=33420/100000 progress=33.4% loss=0.054352 lr=7.949617e-06 elapsed=12:20:17 avg_step=1.33s recent_step=1.53s eta=24:34:49 +2026-08-25 16:15:59 | INFO | train rank=0 | step=33430/100000 progress=33.4% loss=0.044586 lr=7.948282e-06 elapsed=12:20:29 avg_step=1.33s recent_step=1.15s eta=24:34:32 +2026-08-25 16:16:10 | INFO | train rank=0 | step=33440/100000 progress=33.4% loss=0.029671 lr=7.946947e-06 elapsed=12:20:40 avg_step=1.33s recent_step=1.15s eta=24:34:16 +2026-08-25 16:16:22 | INFO | train rank=0 | step=33450/100000 progress=33.5% loss=0.043503 lr=7.945611e-06 elapsed=12:20:52 avg_step=1.33s recent_step=1.16s eta=24:33:59 +2026-08-25 16:16:34 | INFO | train rank=0 | step=33460/100000 progress=33.5% loss=0.060028 lr=7.944274e-06 elapsed=12:21:04 avg_step=1.33s recent_step=1.21s eta=24:33:44 +2026-08-25 16:16:46 | INFO | train rank=0 | step=33470/100000 progress=33.5% loss=0.049286 lr=7.942938e-06 elapsed=12:21:16 avg_step=1.33s recent_step=1.21s eta=24:33:28 +2026-08-25 16:17:01 | INFO | train rank=0 | step=33480/100000 progress=33.5% loss=0.051300 lr=7.941601e-06 elapsed=12:21:30 avg_step=1.33s recent_step=1.43s eta=24:33:17 +2026-08-25 16:17:13 | INFO | train rank=0 | step=33490/100000 progress=33.5% loss=0.067719 lr=7.940264e-06 elapsed=12:21:43 avg_step=1.33s recent_step=1.29s eta=24:33:03 +2026-08-25 16:17:28 | INFO | train rank=0 | step=33500/100000 progress=33.5% loss=0.119629 lr=7.938926e-06 elapsed=12:21:58 avg_step=1.33s recent_step=1.50s eta=24:32:53 +2026-08-25 16:17:40 | INFO | train rank=0 | step=33510/100000 progress=33.5% loss=0.058029 lr=7.937588e-06 elapsed=12:22:10 avg_step=1.33s recent_step=1.16s eta=24:32:36 +2026-08-25 16:17:51 | INFO | train rank=0 | step=33520/100000 progress=33.5% loss=0.038864 lr=7.936250e-06 elapsed=12:22:21 avg_step=1.33s recent_step=1.12s eta=24:32:19 +2026-08-25 16:18:03 | INFO | train rank=0 | step=33530/100000 progress=33.5% loss=0.049789 lr=7.934912e-06 elapsed=12:22:33 avg_step=1.33s recent_step=1.15s eta=24:32:02 +2026-08-25 16:18:18 | INFO | train rank=0 | step=33540/100000 progress=33.5% loss=0.096283 lr=7.933573e-06 elapsed=12:22:47 avg_step=1.33s recent_step=1.47s eta=24:31:51 +2026-08-25 16:18:29 | INFO | train rank=0 | step=33550/100000 progress=33.6% loss=0.050018 lr=7.932234e-06 elapsed=12:22:59 avg_step=1.33s recent_step=1.14s eta=24:31:34 +2026-08-25 16:18:43 | INFO | train rank=0 | step=33560/100000 progress=33.6% loss=0.043060 lr=7.930894e-06 elapsed=12:23:13 avg_step=1.33s recent_step=1.44s eta=24:31:23 +2026-08-25 16:18:55 | INFO | train rank=0 | step=33570/100000 progress=33.6% loss=0.067444 lr=7.929555e-06 elapsed=12:23:25 avg_step=1.33s recent_step=1.20s eta=24:31:07 +2026-08-25 16:19:13 | INFO | train rank=0 | step=33580/100000 progress=33.6% loss=0.057678 lr=7.928215e-06 elapsed=12:23:43 avg_step=1.33s recent_step=1.77s eta=24:31:03 +2026-08-25 16:19:26 | INFO | train rank=0 | step=33590/100000 progress=33.6% loss=0.084595 lr=7.926874e-06 elapsed=12:23:55 avg_step=1.33s recent_step=1.25s eta=24:30:48 +2026-08-25 16:19:41 | INFO | train rank=0 | step=33600/100000 progress=33.6% loss=0.065552 lr=7.925533e-06 elapsed=12:24:11 avg_step=1.33s recent_step=1.56s eta=24:30:39 +2026-08-25 16:19:53 | INFO | train rank=0 | step=33610/100000 progress=33.6% loss=0.062439 lr=7.924192e-06 elapsed=12:24:23 avg_step=1.33s recent_step=1.20s eta=24:30:24 +2026-08-25 16:20:06 | INFO | train rank=0 | step=33620/100000 progress=33.6% loss=0.072266 lr=7.922851e-06 elapsed=12:24:36 avg_step=1.33s recent_step=1.33s eta=24:30:10 +2026-08-25 16:20:21 | INFO | train rank=0 | step=33630/100000 progress=33.6% loss=0.080994 lr=7.921509e-06 elapsed=12:24:51 avg_step=1.33s recent_step=1.44s eta=24:29:59 +2026-08-25 16:20:33 | INFO | train rank=0 | step=33640/100000 progress=33.6% loss=0.058075 lr=7.920167e-06 elapsed=12:25:03 avg_step=1.33s recent_step=1.21s eta=24:29:44 +2026-08-25 16:20:47 | INFO | train rank=0 | step=33650/100000 progress=33.7% loss=0.043106 lr=7.918825e-06 elapsed=12:25:17 avg_step=1.33s recent_step=1.44s eta=24:29:33 +2026-08-25 16:20:59 | INFO | train rank=0 | step=33660/100000 progress=33.7% loss=0.045509 lr=7.917482e-06 elapsed=12:25:29 avg_step=1.33s recent_step=1.16s eta=24:29:16 +2026-08-25 16:21:13 | INFO | train rank=0 | step=33670/100000 progress=33.7% loss=0.031746 lr=7.916139e-06 elapsed=12:25:43 avg_step=1.33s recent_step=1.40s eta=24:29:04 +2026-08-25 16:21:25 | INFO | train rank=0 | step=33680/100000 progress=33.7% loss=0.050705 lr=7.914796e-06 elapsed=12:25:55 avg_step=1.33s recent_step=1.22s eta=24:28:49 +2026-08-25 16:21:37 | INFO | train rank=0 | step=33690/100000 progress=33.7% loss=0.053146 lr=7.913452e-06 elapsed=12:26:07 avg_step=1.33s recent_step=1.20s eta=24:28:33 +2026-08-25 16:21:52 | INFO | train rank=0 | step=33700/100000 progress=33.7% loss=0.045929 lr=7.912108e-06 elapsed=12:26:22 avg_step=1.33s recent_step=1.49s eta=24:28:23 +2026-08-25 16:22:06 | INFO | train rank=0 | step=33710/100000 progress=33.7% loss=0.053284 lr=7.910764e-06 elapsed=12:26:36 avg_step=1.33s recent_step=1.44s eta=24:28:12 +2026-08-25 16:22:24 | INFO | train rank=0 | step=33720/100000 progress=33.7% loss=0.073547 lr=7.909420e-06 elapsed=12:26:54 avg_step=1.33s recent_step=1.72s eta=24:28:06 +2026-08-25 16:22:36 | INFO | train rank=0 | step=33730/100000 progress=33.7% loss=0.046555 lr=7.908075e-06 elapsed=12:27:06 avg_step=1.33s recent_step=1.28s eta=24:27:52 +2026-08-25 16:22:51 | INFO | train rank=0 | step=33740/100000 progress=33.7% loss=0.066803 lr=7.906730e-06 elapsed=12:27:21 avg_step=1.33s recent_step=1.48s eta=24:27:41 +2026-08-25 16:23:03 | INFO | train rank=0 | step=33750/100000 progress=33.8% loss=0.034523 lr=7.905384e-06 elapsed=12:27:32 avg_step=1.33s recent_step=1.14s eta=24:27:24 +2026-08-25 16:23:14 | INFO | train rank=0 | step=33760/100000 progress=33.8% loss=0.040268 lr=7.904038e-06 elapsed=12:27:44 avg_step=1.33s recent_step=1.14s eta=24:27:07 +2026-08-25 16:23:25 | INFO | train rank=0 | step=33770/100000 progress=33.8% loss=0.120667 lr=7.902692e-06 elapsed=12:27:55 avg_step=1.33s recent_step=1.14s eta=24:26:50 +2026-08-25 16:23:41 | INFO | train rank=0 | step=33780/100000 progress=33.8% loss=0.060181 lr=7.901346e-06 elapsed=12:28:11 avg_step=1.33s recent_step=1.56s eta=24:26:42 +2026-08-25 16:23:54 | INFO | train rank=0 | step=33790/100000 progress=33.8% loss=0.090515 lr=7.899999e-06 elapsed=12:28:24 avg_step=1.33s recent_step=1.34s eta=24:26:29 +2026-08-25 16:24:08 | INFO | train rank=0 | step=33800/100000 progress=33.8% loss=0.091919 lr=7.898652e-06 elapsed=12:28:38 avg_step=1.33s recent_step=1.34s eta=24:26:15 +2026-08-25 16:24:21 | INFO | train rank=0 | step=33810/100000 progress=33.8% loss=0.049713 lr=7.897304e-06 elapsed=12:28:51 avg_step=1.33s recent_step=1.28s eta=24:26:01 +2026-08-25 16:24:32 | INFO | train rank=0 | step=33820/100000 progress=33.8% loss=0.041870 lr=7.895957e-06 elapsed=12:29:02 avg_step=1.33s recent_step=1.17s eta=24:25:45 +2026-08-25 16:24:47 | INFO | train rank=0 | step=33830/100000 progress=33.8% loss=0.082581 lr=7.894609e-06 elapsed=12:29:17 avg_step=1.33s recent_step=1.44s eta=24:25:34 +2026-08-25 16:24:58 | INFO | train rank=0 | step=33840/100000 progress=33.8% loss=0.116455 lr=7.893260e-06 elapsed=12:29:28 avg_step=1.33s recent_step=1.15s eta=24:25:17 +2026-08-25 16:25:13 | INFO | train rank=0 | step=33850/100000 progress=33.9% loss=0.039383 lr=7.891911e-06 elapsed=12:29:43 avg_step=1.33s recent_step=1.48s eta=24:25:07 +2026-08-25 16:25:31 | INFO | train rank=0 | step=33860/100000 progress=33.9% loss=0.062439 lr=7.890562e-06 elapsed=12:30:01 avg_step=1.33s recent_step=1.79s eta=24:25:02 +2026-08-25 16:25:43 | INFO | train rank=0 | step=33870/100000 progress=33.9% loss=0.029541 lr=7.889213e-06 elapsed=12:30:13 avg_step=1.33s recent_step=1.21s eta=24:24:47 +2026-08-25 16:25:57 | INFO | train rank=0 | step=33880/100000 progress=33.9% loss=0.058853 lr=7.887864e-06 elapsed=12:30:27 avg_step=1.33s recent_step=1.36s eta=24:24:34 +2026-08-25 16:26:10 | INFO | train rank=0 | step=33890/100000 progress=33.9% loss=0.038429 lr=7.886514e-06 elapsed=12:30:40 avg_step=1.33s recent_step=1.37s eta=24:24:21 +2026-08-25 16:26:22 | INFO | train rank=0 | step=33900/100000 progress=33.9% loss=0.059525 lr=7.885163e-06 elapsed=12:30:52 avg_step=1.33s recent_step=1.19s eta=24:24:05 +2026-08-25 16:26:36 | INFO | train rank=0 | step=33910/100000 progress=33.9% loss=0.049988 lr=7.883813e-06 elapsed=12:31:06 avg_step=1.33s recent_step=1.40s eta=24:23:53 +2026-08-25 16:26:48 | INFO | train rank=0 | step=33920/100000 progress=33.9% loss=0.035652 lr=7.882462e-06 elapsed=12:31:17 avg_step=1.33s recent_step=1.14s eta=24:23:36 +2026-08-25 16:27:00 | INFO | train rank=0 | step=33930/100000 progress=33.9% loss=0.049637 lr=7.881111e-06 elapsed=12:31:30 avg_step=1.33s recent_step=1.21s eta=24:23:21 +2026-08-25 16:27:12 | INFO | train rank=0 | step=33940/100000 progress=33.9% loss=0.040787 lr=7.879759e-06 elapsed=12:31:42 avg_step=1.33s recent_step=1.25s eta=24:23:06 +2026-08-25 16:27:26 | INFO | train rank=0 | step=33950/100000 progress=34.0% loss=0.041245 lr=7.878407e-06 elapsed=12:31:56 avg_step=1.33s recent_step=1.38s eta=24:22:54 +2026-08-25 16:27:41 | INFO | train rank=0 | step=33960/100000 progress=34.0% loss=0.049789 lr=7.877055e-06 elapsed=12:32:11 avg_step=1.33s recent_step=1.54s eta=24:22:45 +2026-08-25 16:27:57 | INFO | train rank=0 | step=33970/100000 progress=34.0% loss=0.041092 lr=7.875703e-06 elapsed=12:32:27 avg_step=1.33s recent_step=1.58s eta=24:22:36 +2026-08-25 16:28:09 | INFO | train rank=0 | step=33980/100000 progress=34.0% loss=0.062256 lr=7.874350e-06 elapsed=12:32:39 avg_step=1.33s recent_step=1.17s eta=24:22:20 +2026-08-25 16:28:20 | INFO | train rank=0 | step=33990/100000 progress=34.0% loss=0.052765 lr=7.872997e-06 elapsed=12:32:50 avg_step=1.33s recent_step=1.15s eta=24:22:03 +2026-08-25 16:28:39 | INFO | train rank=0 | step=34000/100000 progress=34.0% loss=0.035248 lr=7.871643e-06 elapsed=12:33:08 avg_step=1.33s recent_step=1.82s eta=24:21:59 +2026-08-25 16:28:51 | INFO | train rank=0 | step=34010/100000 progress=34.0% loss=0.055115 lr=7.870290e-06 elapsed=12:33:21 avg_step=1.33s recent_step=1.21s eta=24:21:44 +2026-08-25 16:29:05 | INFO | train rank=0 | step=34020/100000 progress=34.0% loss=0.059570 lr=7.868936e-06 elapsed=12:33:35 avg_step=1.33s recent_step=1.45s eta=24:21:33 +2026-08-25 16:29:18 | INFO | train rank=0 | step=34030/100000 progress=34.0% loss=0.042938 lr=7.867581e-06 elapsed=12:33:48 avg_step=1.33s recent_step=1.28s eta=24:21:18 +2026-08-25 16:29:33 | INFO | train rank=0 | step=34040/100000 progress=34.0% loss=0.104675 lr=7.866227e-06 elapsed=12:34:03 avg_step=1.33s recent_step=1.47s eta=24:21:08 +2026-08-25 16:29:47 | INFO | train rank=0 | step=34050/100000 progress=34.1% loss=0.067596 lr=7.864872e-06 elapsed=12:34:17 avg_step=1.33s recent_step=1.48s eta=24:20:57 +2026-08-25 16:30:02 | INFO | train rank=0 | step=34060/100000 progress=34.1% loss=0.053802 lr=7.863516e-06 elapsed=12:34:32 avg_step=1.33s recent_step=1.44s eta=24:20:46 +2026-08-25 16:30:14 | INFO | train rank=0 | step=34070/100000 progress=34.1% loss=0.046509 lr=7.862161e-06 elapsed=12:34:44 avg_step=1.33s recent_step=1.23s eta=24:20:31 +2026-08-25 16:30:30 | INFO | train rank=0 | step=34080/100000 progress=34.1% loss=0.055969 lr=7.860805e-06 elapsed=12:35:00 avg_step=1.33s recent_step=1.57s eta=24:20:22 +2026-08-25 16:30:41 | INFO | train rank=0 | step=34090/100000 progress=34.1% loss=0.034515 lr=7.859448e-06 elapsed=12:35:11 avg_step=1.33s recent_step=1.14s eta=24:20:06 +2026-08-25 16:30:56 | INFO | train rank=0 | step=34100/100000 progress=34.1% loss=0.039108 lr=7.858092e-06 elapsed=12:35:26 avg_step=1.33s recent_step=1.46s eta=24:19:55 +2026-08-25 16:31:07 | INFO | train rank=0 | step=34110/100000 progress=34.1% loss=0.050247 lr=7.856735e-06 elapsed=12:35:37 avg_step=1.33s recent_step=1.15s eta=24:19:38 +2026-08-25 16:31:19 | INFO | train rank=0 | step=34120/100000 progress=34.1% loss=0.059235 lr=7.855378e-06 elapsed=12:35:49 avg_step=1.33s recent_step=1.18s eta=24:19:22 +2026-08-25 16:31:33 | INFO | train rank=0 | step=34130/100000 progress=34.1% loss=0.029053 lr=7.854020e-06 elapsed=12:36:03 avg_step=1.33s recent_step=1.44s eta=24:19:11 +2026-08-25 16:31:48 | INFO | train rank=0 | step=34140/100000 progress=34.1% loss=0.042000 lr=7.852663e-06 elapsed=12:36:18 avg_step=1.33s recent_step=1.44s eta=24:18:59 +2026-08-25 16:32:01 | INFO | train rank=0 | step=34150/100000 progress=34.2% loss=0.044098 lr=7.851304e-06 elapsed=12:36:31 avg_step=1.33s recent_step=1.35s eta=24:18:46 +2026-08-25 16:32:13 | INFO | train rank=0 | step=34160/100000 progress=34.2% loss=0.047607 lr=7.849946e-06 elapsed=12:36:42 avg_step=1.33s recent_step=1.12s eta=24:18:29 +2026-08-25 16:32:28 | INFO | train rank=0 | step=34170/100000 progress=34.2% loss=0.057587 lr=7.848587e-06 elapsed=12:36:58 avg_step=1.33s recent_step=1.54s eta=24:18:20 +2026-08-25 16:32:42 | INFO | train rank=0 | step=34180/100000 progress=34.2% loss=0.043518 lr=7.847228e-06 elapsed=12:37:12 avg_step=1.33s recent_step=1.41s eta=24:18:08 +2026-08-25 16:32:53 | INFO | train rank=0 | step=34190/100000 progress=34.2% loss=0.083679 lr=7.845869e-06 elapsed=12:37:23 avg_step=1.33s recent_step=1.12s eta=24:17:51 +2026-08-25 16:33:08 | INFO | train rank=0 | step=34200/100000 progress=34.2% loss=0.047028 lr=7.844509e-06 elapsed=12:37:38 avg_step=1.33s recent_step=1.44s eta=24:17:40 +2026-08-25 16:33:24 | INFO | train rank=0 | step=34210/100000 progress=34.2% loss=0.047569 lr=7.843149e-06 elapsed=12:37:54 avg_step=1.33s recent_step=1.65s eta=24:17:33 +2026-08-25 16:33:35 | INFO | train rank=0 | step=34220/100000 progress=34.2% loss=0.037281 lr=7.841789e-06 elapsed=12:38:05 avg_step=1.33s recent_step=1.13s eta=24:17:15 +2026-08-25 16:33:50 | INFO | train rank=0 | step=34230/100000 progress=34.2% loss=0.043884 lr=7.840428e-06 elapsed=12:38:20 avg_step=1.33s recent_step=1.45s eta=24:17:05 +2026-08-25 16:34:01 | INFO | train rank=0 | step=34240/100000 progress=34.2% loss=0.072800 lr=7.839067e-06 elapsed=12:38:31 avg_step=1.33s recent_step=1.14s eta=24:16:48 +2026-08-25 16:34:18 | INFO | train rank=0 | step=34250/100000 progress=34.2% loss=0.057114 lr=7.837706e-06 elapsed=12:38:48 avg_step=1.33s recent_step=1.71s eta=24:16:42 +2026-08-25 16:34:30 | INFO | train rank=0 | step=34260/100000 progress=34.3% loss=0.044510 lr=7.836345e-06 elapsed=12:39:00 avg_step=1.33s recent_step=1.16s eta=24:16:25 +2026-08-25 16:34:47 | INFO | train rank=0 | step=34270/100000 progress=34.3% loss=0.049194 lr=7.834983e-06 elapsed=12:39:17 avg_step=1.33s recent_step=1.68s eta=24:16:19 +2026-08-25 16:34:59 | INFO | train rank=0 | step=34280/100000 progress=34.3% loss=0.089447 lr=7.833621e-06 elapsed=12:39:29 avg_step=1.33s recent_step=1.21s eta=24:16:03 +2026-08-25 16:35:10 | INFO | train rank=0 | step=34290/100000 progress=34.3% loss=0.050949 lr=7.832258e-06 elapsed=12:39:40 avg_step=1.33s recent_step=1.10s eta=24:15:45 +2026-08-25 16:35:22 | INFO | train rank=0 | step=34300/100000 progress=34.3% loss=0.081085 lr=7.830896e-06 elapsed=12:39:52 avg_step=1.33s recent_step=1.18s eta=24:15:29 +2026-08-25 16:35:34 | INFO | train rank=0 | step=34310/100000 progress=34.3% loss=0.064453 lr=7.829532e-06 elapsed=12:40:04 avg_step=1.33s recent_step=1.18s eta=24:15:13 +2026-08-25 16:35:46 | INFO | train rank=0 | step=34320/100000 progress=34.3% loss=0.080719 lr=7.828169e-06 elapsed=12:40:16 avg_step=1.33s recent_step=1.22s eta=24:14:58 +2026-08-25 16:35:58 | INFO | train rank=0 | step=34330/100000 progress=34.3% loss=0.043289 lr=7.826805e-06 elapsed=12:40:28 avg_step=1.33s recent_step=1.24s eta=24:14:43 +2026-08-25 16:36:12 | INFO | train rank=0 | step=34340/100000 progress=34.3% loss=0.058746 lr=7.825441e-06 elapsed=12:40:42 avg_step=1.33s recent_step=1.42s eta=24:14:31 +2026-08-25 16:36:27 | INFO | train rank=0 | step=34350/100000 progress=34.4% loss=0.049576 lr=7.824077e-06 elapsed=12:40:57 avg_step=1.33s recent_step=1.43s eta=24:14:20 +2026-08-25 16:36:39 | INFO | train rank=0 | step=34360/100000 progress=34.4% loss=0.054688 lr=7.822712e-06 elapsed=12:41:08 avg_step=1.33s recent_step=1.18s eta=24:14:04 +2026-08-25 16:36:51 | INFO | train rank=0 | step=34370/100000 progress=34.4% loss=0.037399 lr=7.821347e-06 elapsed=12:41:20 avg_step=1.33s recent_step=1.20s eta=24:13:48 +2026-08-25 16:37:05 | INFO | train rank=0 | step=34380/100000 progress=34.4% loss=0.058746 lr=7.819982e-06 elapsed=12:41:35 avg_step=1.33s recent_step=1.42s eta=24:13:36 +2026-08-25 16:37:16 | INFO | train rank=0 | step=34390/100000 progress=34.4% loss=0.066376 lr=7.818617e-06 elapsed=12:41:46 avg_step=1.33s recent_step=1.17s eta=24:13:20 +2026-08-25 16:37:28 | INFO | train rank=0 | step=34400/100000 progress=34.4% loss=0.054932 lr=7.817251e-06 elapsed=12:41:58 avg_step=1.33s recent_step=1.16s eta=24:13:04 +2026-08-25 16:37:42 | INFO | train rank=0 | step=34410/100000 progress=34.4% loss=0.081207 lr=7.815885e-06 elapsed=12:42:12 avg_step=1.33s recent_step=1.42s eta=24:12:52 +2026-08-25 16:37:58 | INFO | train rank=0 | step=34420/100000 progress=34.4% loss=0.046219 lr=7.814518e-06 elapsed=12:42:28 avg_step=1.33s recent_step=1.56s eta=24:12:43 +2026-08-25 16:38:09 | INFO | train rank=0 | step=34430/100000 progress=34.4% loss=0.046722 lr=7.813151e-06 elapsed=12:42:39 avg_step=1.33s recent_step=1.14s eta=24:12:26 +2026-08-25 16:38:28 | INFO | train rank=0 | step=34440/100000 progress=34.4% loss=0.079865 lr=7.811784e-06 elapsed=12:42:58 avg_step=1.33s recent_step=1.83s eta=24:12:23 +2026-08-25 16:38:39 | INFO | train rank=0 | step=34450/100000 progress=34.4% loss=0.102051 lr=7.810417e-06 elapsed=12:43:09 avg_step=1.33s recent_step=1.16s eta=24:12:06 +2026-08-25 16:38:56 | INFO | train rank=0 | step=34460/100000 progress=34.5% loss=0.053497 lr=7.809049e-06 elapsed=12:43:26 avg_step=1.33s recent_step=1.70s eta=24:12:00 +2026-08-25 16:39:11 | INFO | train rank=0 | step=34470/100000 progress=34.5% loss=0.040100 lr=7.807681e-06 elapsed=12:43:41 avg_step=1.33s recent_step=1.47s eta=24:11:49 +2026-08-25 16:39:22 | INFO | train rank=0 | step=34480/100000 progress=34.5% loss=0.087433 lr=7.806313e-06 elapsed=12:43:52 avg_step=1.33s recent_step=1.13s eta=24:11:32 +2026-08-25 16:39:39 | INFO | train rank=0 | step=34490/100000 progress=34.5% loss=0.052338 lr=7.804944e-06 elapsed=12:44:09 avg_step=1.33s recent_step=1.65s eta=24:11:25 +2026-08-25 16:39:51 | INFO | train rank=0 | step=34500/100000 progress=34.5% loss=0.065094 lr=7.803575e-06 elapsed=12:44:20 avg_step=1.33s recent_step=1.17s eta=24:11:09 +2026-08-25 16:40:02 | INFO | train rank=0 | step=34510/100000 progress=34.5% loss=0.033905 lr=7.802206e-06 elapsed=12:44:32 avg_step=1.33s recent_step=1.11s eta=24:10:51 +2026-08-25 16:40:13 | INFO | train rank=0 | step=34520/100000 progress=34.5% loss=0.062622 lr=7.800836e-06 elapsed=12:44:43 avg_step=1.33s recent_step=1.16s eta=24:10:35 +2026-08-25 16:40:25 | INFO | train rank=0 | step=34530/100000 progress=34.5% loss=0.047012 lr=7.799467e-06 elapsed=12:44:55 avg_step=1.33s recent_step=1.15s eta=24:10:18 +2026-08-25 16:40:36 | INFO | train rank=0 | step=34540/100000 progress=34.5% loss=0.065521 lr=7.798096e-06 elapsed=12:45:06 avg_step=1.33s recent_step=1.13s eta=24:10:01 +2026-08-25 16:40:47 | INFO | train rank=0 | step=34550/100000 progress=34.5% loss=0.057419 lr=7.796726e-06 elapsed=12:45:17 avg_step=1.33s recent_step=1.13s eta=24:09:44 +2026-08-25 16:40:59 | INFO | train rank=0 | step=34560/100000 progress=34.6% loss=0.112305 lr=7.795355e-06 elapsed=12:45:28 avg_step=1.33s recent_step=1.12s eta=24:09:27 +2026-08-25 16:41:10 | INFO | train rank=0 | step=34570/100000 progress=34.6% loss=0.068939 lr=7.793984e-06 elapsed=12:45:40 avg_step=1.33s recent_step=1.13s eta=24:09:10 +2026-08-25 16:41:26 | INFO | train rank=0 | step=34580/100000 progress=34.6% loss=0.056183 lr=7.792613e-06 elapsed=12:45:56 avg_step=1.33s recent_step=1.58s eta=24:09:01 +2026-08-25 16:41:37 | INFO | train rank=0 | step=34590/100000 progress=34.6% loss=0.052887 lr=7.791241e-06 elapsed=12:46:07 avg_step=1.33s recent_step=1.13s eta=24:08:44 +2026-08-25 16:41:48 | INFO | train rank=0 | step=34600/100000 progress=34.6% loss=0.047424 lr=7.789869e-06 elapsed=12:46:18 avg_step=1.33s recent_step=1.13s eta=24:08:27 +2026-08-25 16:42:00 | INFO | train rank=0 | step=34610/100000 progress=34.6% loss=0.068787 lr=7.788497e-06 elapsed=12:46:30 avg_step=1.33s recent_step=1.15s eta=24:08:10 +2026-08-25 16:42:18 | INFO | train rank=0 | step=34620/100000 progress=34.6% loss=0.043152 lr=7.787124e-06 elapsed=12:46:48 avg_step=1.33s recent_step=1.84s eta=24:08:07 +2026-08-25 16:42:34 | INFO | train rank=0 | step=34630/100000 progress=34.6% loss=0.050232 lr=7.785751e-06 elapsed=12:47:04 avg_step=1.33s recent_step=1.63s eta=24:07:59 +2026-08-25 16:42:47 | INFO | train rank=0 | step=34640/100000 progress=34.6% loss=0.084381 lr=7.784378e-06 elapsed=12:47:17 avg_step=1.33s recent_step=1.25s eta=24:07:45 +2026-08-25 16:43:01 | INFO | train rank=0 | step=34650/100000 progress=34.6% loss=0.072510 lr=7.783005e-06 elapsed=12:47:31 avg_step=1.33s recent_step=1.36s eta=24:07:32 +2026-08-25 16:43:13 | INFO | train rank=0 | step=34660/100000 progress=34.7% loss=0.065247 lr=7.781631e-06 elapsed=12:47:43 avg_step=1.33s recent_step=1.24s eta=24:07:17 +2026-08-25 16:43:27 | INFO | train rank=0 | step=34670/100000 progress=34.7% loss=0.058472 lr=7.780257e-06 elapsed=12:47:57 avg_step=1.33s recent_step=1.37s eta=24:07:04 +2026-08-25 16:43:39 | INFO | train rank=0 | step=34680/100000 progress=34.7% loss=0.053802 lr=7.778882e-06 elapsed=12:48:09 avg_step=1.33s recent_step=1.21s eta=24:06:49 +2026-08-25 16:43:50 | INFO | train rank=0 | step=34690/100000 progress=34.7% loss=0.046127 lr=7.777507e-06 elapsed=12:48:20 avg_step=1.33s recent_step=1.16s eta=24:06:32 +2026-08-25 16:44:05 | INFO | train rank=0 | step=34700/100000 progress=34.7% loss=0.064209 lr=7.776132e-06 elapsed=12:48:35 avg_step=1.33s recent_step=1.45s eta=24:06:21 +2026-08-25 16:44:18 | INFO | train rank=0 | step=34710/100000 progress=34.7% loss=0.037514 lr=7.774757e-06 elapsed=12:48:48 avg_step=1.33s recent_step=1.33s eta=24:06:08 +2026-08-25 16:44:30 | INFO | train rank=0 | step=34720/100000 progress=34.7% loss=0.035004 lr=7.773381e-06 elapsed=12:49:00 avg_step=1.33s recent_step=1.14s eta=24:05:51 +2026-08-25 16:44:42 | INFO | train rank=0 | step=34730/100000 progress=34.7% loss=0.049271 lr=7.772005e-06 elapsed=12:49:12 avg_step=1.33s recent_step=1.21s eta=24:05:36 +2026-08-25 16:44:57 | INFO | train rank=0 | step=34740/100000 progress=34.7% loss=0.041069 lr=7.770629e-06 elapsed=12:49:27 avg_step=1.33s recent_step=1.52s eta=24:05:26 +2026-08-25 16:45:08 | INFO | train rank=0 | step=34750/100000 progress=34.8% loss=0.048950 lr=7.769253e-06 elapsed=12:49:38 avg_step=1.33s recent_step=1.11s eta=24:05:09 +2026-08-25 16:45:23 | INFO | train rank=0 | step=34760/100000 progress=34.8% loss=0.065674 lr=7.767876e-06 elapsed=12:49:52 avg_step=1.33s recent_step=1.45s eta=24:04:58 +2026-08-25 16:45:35 | INFO | train rank=0 | step=34770/100000 progress=34.8% loss=0.053101 lr=7.766499e-06 elapsed=12:50:05 avg_step=1.33s recent_step=1.22s eta=24:04:42 +2026-08-25 16:45:49 | INFO | train rank=0 | step=34780/100000 progress=34.8% loss=0.081360 lr=7.765121e-06 elapsed=12:50:19 avg_step=1.33s recent_step=1.42s eta=24:04:31 +2026-08-25 16:46:01 | INFO | train rank=0 | step=34790/100000 progress=34.8% loss=0.044708 lr=7.763743e-06 elapsed=12:50:31 avg_step=1.33s recent_step=1.21s eta=24:04:15 +2026-08-25 16:46:15 | INFO | train rank=0 | step=34800/100000 progress=34.8% loss=0.048767 lr=7.762365e-06 elapsed=12:50:45 avg_step=1.33s recent_step=1.43s eta=24:04:04 +2026-08-25 16:46:27 | INFO | train rank=0 | step=34810/100000 progress=34.8% loss=0.051132 lr=7.760987e-06 elapsed=12:50:57 avg_step=1.33s recent_step=1.21s eta=24:03:48 +2026-08-25 16:46:44 | INFO | train rank=0 | step=34820/100000 progress=34.8% loss=0.043472 lr=7.759608e-06 elapsed=12:51:14 avg_step=1.33s recent_step=1.63s eta=24:03:40 +2026-08-25 16:46:57 | INFO | train rank=0 | step=34830/100000 progress=34.8% loss=0.047333 lr=7.758229e-06 elapsed=12:51:27 avg_step=1.33s recent_step=1.34s eta=24:03:27 +2026-08-25 16:47:14 | INFO | train rank=0 | step=34840/100000 progress=34.8% loss=0.052872 lr=7.756850e-06 elapsed=12:51:44 avg_step=1.33s recent_step=1.68s eta=24:03:21 +2026-08-25 16:47:27 | INFO | train rank=0 | step=34850/100000 progress=34.8% loss=0.052246 lr=7.755471e-06 elapsed=12:51:56 avg_step=1.33s recent_step=1.28s eta=24:03:06 +2026-08-25 16:47:39 | INFO | train rank=0 | step=34860/100000 progress=34.9% loss=0.040901 lr=7.754091e-06 elapsed=12:52:09 avg_step=1.33s recent_step=1.23s eta=24:02:51 +2026-08-25 16:47:53 | INFO | train rank=0 | step=34870/100000 progress=34.9% loss=0.076172 lr=7.752710e-06 elapsed=12:52:23 avg_step=1.33s recent_step=1.44s eta=24:02:40 +2026-08-25 16:48:05 | INFO | train rank=0 | step=34880/100000 progress=34.9% loss=0.086853 lr=7.751330e-06 elapsed=12:52:35 avg_step=1.33s recent_step=1.15s eta=24:02:23 +2026-08-25 16:48:19 | INFO | train rank=0 | step=34890/100000 progress=34.9% loss=0.100647 lr=7.749949e-06 elapsed=12:52:49 avg_step=1.33s recent_step=1.41s eta=24:02:12 +2026-08-25 16:48:38 | INFO | train rank=0 | step=34900/100000 progress=34.9% loss=0.065155 lr=7.748568e-06 elapsed=12:53:08 avg_step=1.33s recent_step=1.88s eta=24:02:08 +2026-08-25 16:48:50 | INFO | train rank=0 | step=34910/100000 progress=34.9% loss=0.101868 lr=7.747187e-06 elapsed=12:53:20 avg_step=1.33s recent_step=1.25s eta=24:01:54 +2026-08-25 16:49:01 | INFO | train rank=0 | step=34920/100000 progress=34.9% loss=0.057251 lr=7.745805e-06 elapsed=12:53:31 avg_step=1.33s recent_step=1.12s eta=24:01:37 +2026-08-25 16:49:13 | INFO | train rank=0 | step=34930/100000 progress=34.9% loss=0.118652 lr=7.744423e-06 elapsed=12:53:43 avg_step=1.33s recent_step=1.19s eta=24:01:21 +2026-08-25 16:49:27 | INFO | train rank=0 | step=34940/100000 progress=34.9% loss=0.044296 lr=7.743041e-06 elapsed=12:53:57 avg_step=1.33s recent_step=1.38s eta=24:01:08 +2026-08-25 16:49:39 | INFO | train rank=0 | step=34950/100000 progress=34.9% loss=0.039375 lr=7.741658e-06 elapsed=12:54:09 avg_step=1.33s recent_step=1.18s eta=24:00:52 +2026-08-25 16:49:55 | INFO | train rank=0 | step=34960/100000 progress=35.0% loss=0.083740 lr=7.740275e-06 elapsed=12:54:24 avg_step=1.33s recent_step=1.56s eta=24:00:43 +2026-08-25 16:50:07 | INFO | train rank=0 | step=34970/100000 progress=35.0% loss=0.049805 lr=7.738892e-06 elapsed=12:54:37 avg_step=1.33s recent_step=1.23s eta=24:00:28 +2026-08-25 16:50:23 | INFO | train rank=0 | step=34980/100000 progress=35.0% loss=0.055588 lr=7.737509e-06 elapsed=12:54:53 avg_step=1.33s recent_step=1.58s eta=24:00:20 +2026-08-25 16:50:34 | INFO | train rank=0 | step=34990/100000 progress=35.0% loss=0.048615 lr=7.736125e-06 elapsed=12:55:04 avg_step=1.33s recent_step=1.14s eta=24:00:03 +2026-08-25 16:50:47 | INFO | train rank=0 | step=35000/100000 progress=35.0% loss=0.045319 lr=7.734741e-06 elapsed=12:55:17 avg_step=1.33s recent_step=1.31s eta=23:59:49 +2026-08-25 16:51:00 | INFO | train rank=0 | step=35010/100000 progress=35.0% loss=0.066620 lr=7.733356e-06 elapsed=12:55:30 avg_step=1.33s recent_step=1.29s eta=23:59:35 +2026-08-25 16:51:19 | INFO | train rank=0 | step=35020/100000 progress=35.0% loss=0.036530 lr=7.731972e-06 elapsed=12:55:49 avg_step=1.33s recent_step=1.91s eta=23:59:33 +2026-08-25 16:51:31 | INFO | train rank=0 | step=35030/100000 progress=35.0% loss=0.063766 lr=7.730587e-06 elapsed=12:56:01 avg_step=1.33s recent_step=1.16s eta=23:59:17 +2026-08-25 16:51:42 | INFO | train rank=0 | step=35040/100000 progress=35.0% loss=0.043854 lr=7.729201e-06 elapsed=12:56:12 avg_step=1.33s recent_step=1.12s eta=23:58:59 +2026-08-25 16:51:55 | INFO | train rank=0 | step=35050/100000 progress=35.0% loss=0.055069 lr=7.727816e-06 elapsed=12:56:25 avg_step=1.33s recent_step=1.31s eta=23:58:46 +2026-08-25 16:52:07 | INFO | train rank=0 | step=35060/100000 progress=35.1% loss=0.106079 lr=7.726430e-06 elapsed=12:56:37 avg_step=1.33s recent_step=1.17s eta=23:58:29 +2026-08-25 16:52:22 | INFO | train rank=0 | step=35070/100000 progress=35.1% loss=0.059265 lr=7.725044e-06 elapsed=12:56:52 avg_step=1.33s recent_step=1.48s eta=23:58:19 +2026-08-25 16:52:33 | INFO | train rank=0 | step=35080/100000 progress=35.1% loss=0.079132 lr=7.723657e-06 elapsed=12:57:03 avg_step=1.33s recent_step=1.15s eta=23:58:02 +2026-08-25 16:52:48 | INFO | train rank=0 | step=35090/100000 progress=35.1% loss=0.059891 lr=7.722271e-06 elapsed=12:57:18 avg_step=1.33s recent_step=1.48s eta=23:57:52 +2026-08-25 16:53:01 | INFO | train rank=0 | step=35100/100000 progress=35.1% loss=0.050827 lr=7.720884e-06 elapsed=12:57:31 avg_step=1.33s recent_step=1.27s eta=23:57:38 +2026-08-25 16:53:13 | INFO | train rank=0 | step=35110/100000 progress=35.1% loss=0.044846 lr=7.719496e-06 elapsed=12:57:42 avg_step=1.33s recent_step=1.17s eta=23:57:21 +2026-08-25 16:53:25 | INFO | train rank=0 | step=35120/100000 progress=35.1% loss=0.071289 lr=7.718109e-06 elapsed=12:57:55 avg_step=1.33s recent_step=1.24s eta=23:57:06 +2026-08-25 16:53:38 | INFO | train rank=0 | step=35130/100000 progress=35.1% loss=0.034721 lr=7.716721e-06 elapsed=12:58:08 avg_step=1.33s recent_step=1.28s eta=23:56:52 +2026-08-25 16:53:55 | INFO | train rank=0 | step=35140/100000 progress=35.1% loss=0.029366 lr=7.715332e-06 elapsed=12:58:25 avg_step=1.33s recent_step=1.77s eta=23:56:47 +2026-08-25 16:54:07 | INFO | train rank=0 | step=35150/100000 progress=35.1% loss=0.072418 lr=7.713944e-06 elapsed=12:58:37 avg_step=1.33s recent_step=1.15s eta=23:56:31 +2026-08-25 16:54:18 | INFO | train rank=0 | step=35160/100000 progress=35.2% loss=0.034454 lr=7.712555e-06 elapsed=12:58:48 avg_step=1.33s recent_step=1.16s eta=23:56:14 +2026-08-25 16:54:30 | INFO | train rank=0 | step=35170/100000 progress=35.2% loss=0.071198 lr=7.711166e-06 elapsed=12:59:00 avg_step=1.33s recent_step=1.19s eta=23:55:58 +2026-08-25 16:54:44 | INFO | train rank=0 | step=35180/100000 progress=35.2% loss=0.066742 lr=7.709776e-06 elapsed=12:59:14 avg_step=1.33s recent_step=1.38s eta=23:55:46 +2026-08-25 16:55:00 | INFO | train rank=0 | step=35190/100000 progress=35.2% loss=0.067749 lr=7.708387e-06 elapsed=12:59:29 avg_step=1.33s recent_step=1.53s eta=23:55:36 +2026-08-25 16:55:11 | INFO | train rank=0 | step=35200/100000 progress=35.2% loss=0.038498 lr=7.706997e-06 elapsed=12:59:41 avg_step=1.33s recent_step=1.18s eta=23:55:20 +2026-08-25 16:55:23 | INFO | train rank=0 | step=35210/100000 progress=35.2% loss=0.053520 lr=7.705606e-06 elapsed=12:59:53 avg_step=1.33s recent_step=1.17s eta=23:55:04 +2026-08-25 16:55:38 | INFO | train rank=0 | step=35220/100000 progress=35.2% loss=0.063309 lr=7.704216e-06 elapsed=13:00:08 avg_step=1.33s recent_step=1.53s eta=23:54:55 +2026-08-25 16:55:51 | INFO | train rank=0 | step=35230/100000 progress=35.2% loss=0.049515 lr=7.702825e-06 elapsed=13:00:21 avg_step=1.33s recent_step=1.31s eta=23:54:41 +2026-08-25 16:56:06 | INFO | train rank=0 | step=35240/100000 progress=35.2% loss=0.065430 lr=7.701434e-06 elapsed=13:00:36 avg_step=1.33s recent_step=1.50s eta=23:54:31 +2026-08-25 16:56:18 | INFO | train rank=0 | step=35250/100000 progress=35.2% loss=0.054520 lr=7.700042e-06 elapsed=13:00:48 avg_step=1.33s recent_step=1.17s eta=23:54:15 +2026-08-25 16:56:33 | INFO | train rank=0 | step=35260/100000 progress=35.3% loss=0.047348 lr=7.698650e-06 elapsed=13:01:03 avg_step=1.33s recent_step=1.49s eta=23:54:04 +2026-08-25 16:56:44 | INFO | train rank=0 | step=35270/100000 progress=35.3% loss=0.055222 lr=7.697258e-06 elapsed=13:01:14 avg_step=1.33s recent_step=1.12s eta=23:53:47 +2026-08-25 16:57:01 | INFO | train rank=0 | step=35280/100000 progress=35.3% loss=0.058136 lr=7.695866e-06 elapsed=13:01:31 avg_step=1.33s recent_step=1.72s eta=23:53:41 +2026-08-25 16:57:13 | INFO | train rank=0 | step=35290/100000 progress=35.3% loss=0.057465 lr=7.694473e-06 elapsed=13:01:43 avg_step=1.33s recent_step=1.18s eta=23:53:25 +2026-08-25 16:57:25 | INFO | train rank=0 | step=35300/100000 progress=35.3% loss=0.035034 lr=7.693080e-06 elapsed=13:01:55 avg_step=1.33s recent_step=1.18s eta=23:53:09 +2026-08-25 16:57:37 | INFO | train rank=0 | step=35310/100000 progress=35.3% loss=0.078552 lr=7.691687e-06 elapsed=13:02:07 avg_step=1.33s recent_step=1.16s eta=23:52:53 +2026-08-25 16:57:51 | INFO | train rank=0 | step=35320/100000 progress=35.3% loss=0.060120 lr=7.690293e-06 elapsed=13:02:21 avg_step=1.33s recent_step=1.45s eta=23:52:42 +2026-08-25 16:58:03 | INFO | train rank=0 | step=35330/100000 progress=35.3% loss=0.072388 lr=7.688899e-06 elapsed=13:02:33 avg_step=1.33s recent_step=1.19s eta=23:52:26 +2026-08-25 16:58:16 | INFO | train rank=0 | step=35340/100000 progress=35.3% loss=0.069824 lr=7.687505e-06 elapsed=13:02:46 avg_step=1.33s recent_step=1.30s eta=23:52:12 +2026-08-25 16:58:32 | INFO | train rank=0 | step=35350/100000 progress=35.4% loss=0.078369 lr=7.686111e-06 elapsed=13:03:02 avg_step=1.33s recent_step=1.60s eta=23:52:04 +2026-08-25 16:58:44 | INFO | train rank=0 | step=35360/100000 progress=35.4% loss=0.158875 lr=7.684716e-06 elapsed=13:03:14 avg_step=1.33s recent_step=1.23s eta=23:51:48 +2026-08-25 16:58:58 | INFO | train rank=0 | step=35370/100000 progress=35.4% loss=0.072083 lr=7.683321e-06 elapsed=13:03:28 avg_step=1.33s recent_step=1.38s eta=23:51:36 +2026-08-25 16:59:10 | INFO | train rank=0 | step=35380/100000 progress=35.4% loss=0.043938 lr=7.681926e-06 elapsed=13:03:39 avg_step=1.33s recent_step=1.14s eta=23:51:19 +2026-08-25 16:59:28 | INFO | train rank=0 | step=35390/100000 progress=35.4% loss=0.044022 lr=7.680530e-06 elapsed=13:03:58 avg_step=1.33s recent_step=1.81s eta=23:51:15 +2026-08-25 16:59:39 | INFO | train rank=0 | step=35400/100000 progress=35.4% loss=0.043381 lr=7.679134e-06 elapsed=13:04:09 avg_step=1.33s recent_step=1.17s eta=23:50:59 +2026-08-25 16:59:54 | INFO | train rank=0 | step=35410/100000 progress=35.4% loss=0.057358 lr=7.677738e-06 elapsed=13:04:24 avg_step=1.33s recent_step=1.47s eta=23:50:48 +2026-08-25 17:00:06 | INFO | train rank=0 | step=35420/100000 progress=35.4% loss=0.073273 lr=7.676341e-06 elapsed=13:04:36 avg_step=1.33s recent_step=1.17s eta=23:50:32 +2026-08-25 17:00:19 | INFO | train rank=0 | step=35430/100000 progress=35.4% loss=0.056381 lr=7.674944e-06 elapsed=13:04:49 avg_step=1.33s recent_step=1.30s eta=23:50:18 +2026-08-25 17:00:34 | INFO | train rank=0 | step=35440/100000 progress=35.4% loss=0.034561 lr=7.673547e-06 elapsed=13:05:04 avg_step=1.33s recent_step=1.57s eta=23:50:09 +2026-08-25 17:00:50 | INFO | train rank=0 | step=35450/100000 progress=35.4% loss=0.058151 lr=7.672150e-06 elapsed=13:05:19 avg_step=1.33s recent_step=1.52s eta=23:49:59 +2026-08-25 17:01:01 | INFO | train rank=0 | step=35460/100000 progress=35.5% loss=0.048157 lr=7.670752e-06 elapsed=13:05:31 avg_step=1.33s recent_step=1.13s eta=23:49:42 +2026-08-25 17:01:15 | INFO | train rank=0 | step=35470/100000 progress=35.5% loss=0.058792 lr=7.669354e-06 elapsed=13:05:45 avg_step=1.33s recent_step=1.45s eta=23:49:31 +2026-08-25 17:01:30 | INFO | train rank=0 | step=35480/100000 progress=35.5% loss=0.048676 lr=7.667956e-06 elapsed=13:06:00 avg_step=1.33s recent_step=1.45s eta=23:49:20 +2026-08-25 17:01:42 | INFO | train rank=0 | step=35490/100000 progress=35.5% loss=0.082123 lr=7.666557e-06 elapsed=13:06:11 avg_step=1.33s recent_step=1.16s eta=23:49:04 +2026-08-25 17:01:54 | INFO | train rank=0 | step=35500/100000 progress=35.5% loss=0.061279 lr=7.665159e-06 elapsed=13:06:24 avg_step=1.33s recent_step=1.24s eta=23:48:49 +2026-08-25 17:02:05 | INFO | train rank=0 | step=35510/100000 progress=35.5% loss=0.063507 lr=7.663760e-06 elapsed=13:06:35 avg_step=1.33s recent_step=1.15s eta=23:48:32 +2026-08-25 17:02:19 | INFO | train rank=0 | step=35520/100000 progress=35.5% loss=0.045124 lr=7.662360e-06 elapsed=13:06:49 avg_step=1.33s recent_step=1.35s eta=23:48:19 +2026-08-25 17:02:34 | INFO | train rank=0 | step=35530/100000 progress=35.5% loss=0.040298 lr=7.660960e-06 elapsed=13:07:04 avg_step=1.33s recent_step=1.53s eta=23:48:10 +2026-08-25 17:02:50 | INFO | train rank=0 | step=35540/100000 progress=35.5% loss=0.052231 lr=7.659560e-06 elapsed=13:07:20 avg_step=1.33s recent_step=1.61s eta=23:48:01 +2026-08-25 17:03:06 | INFO | train rank=0 | step=35550/100000 progress=35.5% loss=0.049622 lr=7.658160e-06 elapsed=13:07:36 avg_step=1.33s recent_step=1.56s eta=23:47:52 +2026-08-25 17:03:20 | INFO | train rank=0 | step=35560/100000 progress=35.6% loss=0.047211 lr=7.656759e-06 elapsed=13:07:50 avg_step=1.33s recent_step=1.41s eta=23:47:40 +2026-08-25 17:03:32 | INFO | train rank=0 | step=35570/100000 progress=35.6% loss=0.052673 lr=7.655359e-06 elapsed=13:08:01 avg_step=1.33s recent_step=1.17s eta=23:47:24 +2026-08-25 17:03:43 | INFO | train rank=0 | step=35580/100000 progress=35.6% loss=0.048355 lr=7.653957e-06 elapsed=13:08:13 avg_step=1.33s recent_step=1.15s eta=23:47:08 +2026-08-25 17:03:58 | INFO | train rank=0 | step=35590/100000 progress=35.6% loss=0.073730 lr=7.652556e-06 elapsed=13:08:28 avg_step=1.33s recent_step=1.53s eta=23:46:58 +2026-08-25 17:04:12 | INFO | train rank=0 | step=35600/100000 progress=35.6% loss=0.047173 lr=7.651154e-06 elapsed=13:08:42 avg_step=1.33s recent_step=1.36s eta=23:46:45 +2026-08-25 17:04:24 | INFO | train rank=0 | step=35610/100000 progress=35.6% loss=0.053162 lr=7.649752e-06 elapsed=13:08:54 avg_step=1.33s recent_step=1.19s eta=23:46:29 +2026-08-25 17:04:42 | INFO | train rank=0 | step=35620/100000 progress=35.6% loss=0.057678 lr=7.648350e-06 elapsed=13:09:12 avg_step=1.33s recent_step=1.85s eta=23:46:25 +2026-08-25 17:04:54 | INFO | train rank=0 | step=35630/100000 progress=35.6% loss=0.044464 lr=7.646947e-06 elapsed=13:09:24 avg_step=1.33s recent_step=1.16s eta=23:46:09 +2026-08-25 17:05:10 | INFO | train rank=0 | step=35640/100000 progress=35.6% loss=0.066696 lr=7.645544e-06 elapsed=13:09:40 avg_step=1.33s recent_step=1.64s eta=23:46:01 +2026-08-25 17:05:22 | INFO | train rank=0 | step=35650/100000 progress=35.6% loss=0.079315 lr=7.644141e-06 elapsed=13:09:52 avg_step=1.33s recent_step=1.14s eta=23:45:45 +2026-08-25 17:05:34 | INFO | train rank=0 | step=35660/100000 progress=35.7% loss=0.052689 lr=7.642738e-06 elapsed=13:10:04 avg_step=1.33s recent_step=1.24s eta=23:45:30 +2026-08-25 17:05:49 | INFO | train rank=0 | step=35670/100000 progress=35.7% loss=0.078186 lr=7.641334e-06 elapsed=13:10:19 avg_step=1.33s recent_step=1.47s eta=23:45:19 +2026-08-25 17:06:04 | INFO | train rank=0 | step=35680/100000 progress=35.7% loss=0.064194 lr=7.639930e-06 elapsed=13:10:34 avg_step=1.33s recent_step=1.52s eta=23:45:09 +2026-08-25 17:06:16 | INFO | train rank=0 | step=35690/100000 progress=35.7% loss=0.045990 lr=7.638525e-06 elapsed=13:10:46 avg_step=1.33s recent_step=1.19s eta=23:44:53 +2026-08-25 17:06:28 | INFO | train rank=0 | step=35700/100000 progress=35.7% loss=0.058914 lr=7.637121e-06 elapsed=13:10:58 avg_step=1.33s recent_step=1.21s eta=23:44:38 +2026-08-25 17:06:39 | INFO | train rank=0 | step=35710/100000 progress=35.7% loss=0.058685 lr=7.635716e-06 elapsed=13:11:09 avg_step=1.33s recent_step=1.12s eta=23:44:21 +2026-08-25 17:06:57 | INFO | train rank=0 | step=35720/100000 progress=35.7% loss=0.078217 lr=7.634311e-06 elapsed=13:11:27 avg_step=1.33s recent_step=1.77s eta=23:44:15 +2026-08-25 17:07:09 | INFO | train rank=0 | step=35730/100000 progress=35.7% loss=0.038025 lr=7.632905e-06 elapsed=13:11:39 avg_step=1.33s recent_step=1.18s eta=23:43:59 +2026-08-25 17:07:21 | INFO | train rank=0 | step=35740/100000 progress=35.7% loss=0.067352 lr=7.631499e-06 elapsed=13:11:50 avg_step=1.33s recent_step=1.19s eta=23:43:44 +2026-08-25 17:07:32 | INFO | train rank=0 | step=35750/100000 progress=35.8% loss=0.049759 lr=7.630093e-06 elapsed=13:12:02 avg_step=1.33s recent_step=1.17s eta=23:43:27 +2026-08-25 17:07:48 | INFO | train rank=0 | step=35760/100000 progress=35.8% loss=0.070068 lr=7.628687e-06 elapsed=13:12:18 avg_step=1.33s recent_step=1.61s eta=23:43:19 +2026-08-25 17:08:02 | INFO | train rank=0 | step=35770/100000 progress=35.8% loss=0.064301 lr=7.627280e-06 elapsed=13:12:32 avg_step=1.33s recent_step=1.35s eta=23:43:06 +2026-08-25 17:08:14 | INFO | train rank=0 | step=35780/100000 progress=35.8% loss=0.038223 lr=7.625873e-06 elapsed=13:12:44 avg_step=1.33s recent_step=1.19s eta=23:42:50 +2026-08-25 17:08:27 | INFO | train rank=0 | step=35790/100000 progress=35.8% loss=0.044952 lr=7.624466e-06 elapsed=13:12:57 avg_step=1.33s recent_step=1.33s eta=23:42:37 +2026-08-25 17:08:39 | INFO | train rank=0 | step=35800/100000 progress=35.8% loss=0.056244 lr=7.623058e-06 elapsed=13:13:09 avg_step=1.33s recent_step=1.21s eta=23:42:22 +2026-08-25 17:08:51 | INFO | train rank=0 | step=35810/100000 progress=35.8% loss=0.057281 lr=7.621651e-06 elapsed=13:13:21 avg_step=1.33s recent_step=1.18s eta=23:42:06 +2026-08-25 17:09:06 | INFO | train rank=0 | step=35820/100000 progress=35.8% loss=0.045303 lr=7.620242e-06 elapsed=13:13:36 avg_step=1.33s recent_step=1.49s eta=23:41:55 +2026-08-25 17:09:23 | INFO | train rank=0 | step=35830/100000 progress=35.8% loss=0.041199 lr=7.618834e-06 elapsed=13:13:52 avg_step=1.33s recent_step=1.67s eta=23:41:48 +2026-08-25 17:09:34 | INFO | train rank=0 | step=35840/100000 progress=35.8% loss=0.082642 lr=7.617425e-06 elapsed=13:14:04 avg_step=1.33s recent_step=1.15s eta=23:41:31 +2026-08-25 17:09:46 | INFO | train rank=0 | step=35850/100000 progress=35.9% loss=0.038788 lr=7.616016e-06 elapsed=13:14:15 avg_step=1.33s recent_step=1.16s eta=23:41:15 +2026-08-25 17:09:58 | INFO | train rank=0 | step=35860/100000 progress=35.9% loss=0.036865 lr=7.614607e-06 elapsed=13:14:28 avg_step=1.33s recent_step=1.22s eta=23:41:00 +2026-08-25 17:10:12 | INFO | train rank=0 | step=35870/100000 progress=35.9% loss=0.045929 lr=7.613198e-06 elapsed=13:14:42 avg_step=1.33s recent_step=1.47s eta=23:40:49 +2026-08-25 17:10:24 | INFO | train rank=0 | step=35880/100000 progress=35.9% loss=0.075531 lr=7.611788e-06 elapsed=13:14:54 avg_step=1.33s recent_step=1.17s eta=23:40:33 +2026-08-25 17:10:36 | INFO | train rank=0 | step=35890/100000 progress=35.9% loss=0.061584 lr=7.610378e-06 elapsed=13:15:06 avg_step=1.33s recent_step=1.16s eta=23:40:17 +2026-08-25 17:10:53 | INFO | train rank=0 | step=35900/100000 progress=35.9% loss=0.058807 lr=7.608967e-06 elapsed=13:15:23 avg_step=1.33s recent_step=1.76s eta=23:40:11 +2026-08-25 17:11:05 | INFO | train rank=0 | step=35910/100000 progress=35.9% loss=0.075882 lr=7.607557e-06 elapsed=13:15:35 avg_step=1.33s recent_step=1.20s eta=23:39:55 +2026-08-25 17:11:20 | INFO | train rank=0 | step=35920/100000 progress=35.9% loss=0.041145 lr=7.606146e-06 elapsed=13:15:50 avg_step=1.33s recent_step=1.48s eta=23:39:45 +2026-08-25 17:11:36 | INFO | train rank=0 | step=35930/100000 progress=35.9% loss=0.053314 lr=7.604735e-06 elapsed=13:16:05 avg_step=1.33s recent_step=1.54s eta=23:39:35 +2026-08-25 17:11:47 | INFO | train rank=0 | step=35940/100000 progress=35.9% loss=0.062378 lr=7.603323e-06 elapsed=13:16:17 avg_step=1.33s recent_step=1.19s eta=23:39:19 +2026-08-25 17:12:01 | INFO | train rank=0 | step=35950/100000 progress=35.9% loss=0.045624 lr=7.601911e-06 elapsed=13:16:31 avg_step=1.33s recent_step=1.33s eta=23:39:06 +2026-08-25 17:12:19 | INFO | train rank=0 | step=35960/100000 progress=36.0% loss=0.074829 lr=7.600499e-06 elapsed=13:16:49 avg_step=1.33s recent_step=1.81s eta=23:39:01 +2026-08-25 17:12:30 | INFO | train rank=0 | step=35970/100000 progress=36.0% loss=0.103821 lr=7.599087e-06 elapsed=13:17:00 avg_step=1.33s recent_step=1.16s eta=23:38:45 +2026-08-25 17:12:43 | INFO | train rank=0 | step=35980/100000 progress=36.0% loss=0.038956 lr=7.597674e-06 elapsed=13:17:13 avg_step=1.33s recent_step=1.23s eta=23:38:30 +2026-08-25 17:12:56 | INFO | train rank=0 | step=35990/100000 progress=36.0% loss=0.040413 lr=7.596261e-06 elapsed=13:17:26 avg_step=1.33s recent_step=1.31s eta=23:38:16 +2026-08-25 17:13:09 | INFO | train rank=0 | step=36000/100000 progress=36.0% loss=0.045273 lr=7.594848e-06 elapsed=13:17:39 avg_step=1.33s recent_step=1.31s eta=23:38:03 +2026-08-25 17:13:23 | INFO | train rank=0 | step=36010/100000 progress=36.0% loss=0.043839 lr=7.593434e-06 elapsed=13:17:53 avg_step=1.33s recent_step=1.43s eta=23:37:51 +2026-08-25 17:13:35 | INFO | train rank=0 | step=36020/100000 progress=36.0% loss=0.063202 lr=7.592021e-06 elapsed=13:18:05 avg_step=1.33s recent_step=1.18s eta=23:37:35 +2026-08-25 17:13:47 | INFO | train rank=0 | step=36030/100000 progress=36.0% loss=0.049622 lr=7.590606e-06 elapsed=13:18:17 avg_step=1.33s recent_step=1.19s eta=23:37:20 +2026-08-25 17:14:02 | INFO | train rank=0 | step=36040/100000 progress=36.0% loss=0.078705 lr=7.589192e-06 elapsed=13:18:32 avg_step=1.33s recent_step=1.53s eta=23:37:10 +2026-08-25 17:14:14 | INFO | train rank=0 | step=36050/100000 progress=36.0% loss=0.061768 lr=7.587777e-06 elapsed=13:18:44 avg_step=1.33s recent_step=1.16s eta=23:36:54 +2026-08-25 17:14:25 | INFO | train rank=0 | step=36060/100000 progress=36.1% loss=0.056519 lr=7.586363e-06 elapsed=13:18:55 avg_step=1.33s recent_step=1.14s eta=23:36:37 +2026-08-25 17:14:37 | INFO | train rank=0 | step=36070/100000 progress=36.1% loss=0.064072 lr=7.584947e-06 elapsed=13:19:07 avg_step=1.33s recent_step=1.20s eta=23:36:21 +2026-08-25 17:14:55 | INFO | train rank=0 | step=36080/100000 progress=36.1% loss=0.085114 lr=7.583532e-06 elapsed=13:19:25 avg_step=1.33s recent_step=1.78s eta=23:36:16 +2026-08-25 17:15:07 | INFO | train rank=0 | step=36090/100000 progress=36.1% loss=0.045845 lr=7.582116e-06 elapsed=13:19:37 avg_step=1.33s recent_step=1.18s eta=23:36:00 +2026-08-25 17:15:23 | INFO | train rank=0 | step=36100/100000 progress=36.1% loss=0.053177 lr=7.580700e-06 elapsed=13:19:53 avg_step=1.33s recent_step=1.60s eta=23:35:51 +2026-08-25 17:15:35 | INFO | train rank=0 | step=36110/100000 progress=36.1% loss=0.064316 lr=7.579284e-06 elapsed=13:20:05 avg_step=1.33s recent_step=1.19s eta=23:35:36 +2026-08-25 17:15:52 | INFO | train rank=0 | step=36120/100000 progress=36.1% loss=0.057785 lr=7.577867e-06 elapsed=13:20:22 avg_step=1.33s recent_step=1.74s eta=23:35:30 +2026-08-25 17:16:07 | INFO | train rank=0 | step=36130/100000 progress=36.1% loss=0.065460 lr=7.576450e-06 elapsed=13:20:37 avg_step=1.33s recent_step=1.49s eta=23:35:19 +2026-08-25 17:16:22 | INFO | train rank=0 | step=36140/100000 progress=36.1% loss=0.047516 lr=7.575033e-06 elapsed=13:20:52 avg_step=1.33s recent_step=1.50s eta=23:35:09 +2026-08-25 17:16:33 | INFO | train rank=0 | step=36150/100000 progress=36.1% loss=0.043961 lr=7.573615e-06 elapsed=13:21:03 avg_step=1.33s recent_step=1.15s eta=23:34:52 +2026-08-25 17:16:48 | INFO | train rank=0 | step=36160/100000 progress=36.2% loss=0.077972 lr=7.572198e-06 elapsed=13:21:18 avg_step=1.33s recent_step=1.50s eta=23:34:42 +2026-08-25 17:17:02 | INFO | train rank=0 | step=36170/100000 progress=36.2% loss=0.048920 lr=7.570780e-06 elapsed=13:21:32 avg_step=1.33s recent_step=1.38s eta=23:34:29 +2026-08-25 17:17:15 | INFO | train rank=0 | step=36180/100000 progress=36.2% loss=0.050812 lr=7.569361e-06 elapsed=13:21:45 avg_step=1.33s recent_step=1.29s eta=23:34:15 +2026-08-25 17:17:26 | INFO | train rank=0 | step=36190/100000 progress=36.2% loss=0.054459 lr=7.567943e-06 elapsed=13:21:56 avg_step=1.33s recent_step=1.13s eta=23:33:59 +2026-08-25 17:17:38 | INFO | train rank=0 | step=36200/100000 progress=36.2% loss=0.082214 lr=7.566524e-06 elapsed=13:22:08 avg_step=1.33s recent_step=1.16s eta=23:33:42 +2026-08-25 17:17:51 | INFO | train rank=0 | step=36210/100000 progress=36.2% loss=0.049362 lr=7.565105e-06 elapsed=13:22:20 avg_step=1.33s recent_step=1.27s eta=23:33:28 +2026-08-25 17:18:02 | INFO | train rank=0 | step=36220/100000 progress=36.2% loss=0.041718 lr=7.563685e-06 elapsed=13:22:32 avg_step=1.33s recent_step=1.14s eta=23:33:11 +2026-08-25 17:18:16 | INFO | train rank=0 | step=36230/100000 progress=36.2% loss=0.047318 lr=7.562266e-06 elapsed=13:22:46 avg_step=1.33s recent_step=1.36s eta=23:32:59 +2026-08-25 17:18:30 | INFO | train rank=0 | step=36240/100000 progress=36.2% loss=0.045273 lr=7.560846e-06 elapsed=13:23:00 avg_step=1.33s recent_step=1.47s eta=23:32:48 +2026-08-25 17:18:42 | INFO | train rank=0 | step=36250/100000 progress=36.2% loss=0.099182 lr=7.559425e-06 elapsed=13:23:11 avg_step=1.33s recent_step=1.13s eta=23:32:31 +2026-08-25 17:18:57 | INFO | train rank=0 | step=36260/100000 progress=36.3% loss=0.061386 lr=7.558005e-06 elapsed=13:23:27 avg_step=1.33s recent_step=1.56s eta=23:32:22 +2026-08-25 17:19:10 | INFO | train rank=0 | step=36270/100000 progress=36.3% loss=0.044052 lr=7.556584e-06 elapsed=13:23:40 avg_step=1.33s recent_step=1.33s eta=23:32:08 +2026-08-25 17:19:22 | INFO | train rank=0 | step=36280/100000 progress=36.3% loss=0.076172 lr=7.555163e-06 elapsed=13:23:52 avg_step=1.33s recent_step=1.17s eta=23:31:52 +2026-08-25 17:19:36 | INFO | train rank=0 | step=36290/100000 progress=36.3% loss=0.063141 lr=7.553741e-06 elapsed=13:24:05 avg_step=1.33s recent_step=1.34s eta=23:31:39 +2026-08-25 17:19:47 | INFO | train rank=0 | step=36300/100000 progress=36.3% loss=0.054672 lr=7.552320e-06 elapsed=13:24:17 avg_step=1.33s recent_step=1.16s eta=23:31:23 +2026-08-25 17:19:59 | INFO | train rank=0 | step=36310/100000 progress=36.3% loss=0.061615 lr=7.550898e-06 elapsed=13:24:29 avg_step=1.33s recent_step=1.17s eta=23:31:07 +2026-08-25 17:20:11 | INFO | train rank=0 | step=36320/100000 progress=36.3% loss=0.060852 lr=7.549475e-06 elapsed=13:24:41 avg_step=1.33s recent_step=1.20s eta=23:30:51 +2026-08-25 17:20:23 | INFO | train rank=0 | step=36330/100000 progress=36.3% loss=0.051926 lr=7.548053e-06 elapsed=13:24:53 avg_step=1.33s recent_step=1.25s eta=23:30:37 +2026-08-25 17:20:35 | INFO | train rank=0 | step=36340/100000 progress=36.3% loss=0.048462 lr=7.546630e-06 elapsed=13:25:05 avg_step=1.33s recent_step=1.21s eta=23:30:21 +2026-08-25 17:20:47 | INFO | train rank=0 | step=36350/100000 progress=36.4% loss=0.050491 lr=7.545207e-06 elapsed=13:25:17 avg_step=1.33s recent_step=1.17s eta=23:30:05 +2026-08-25 17:20:59 | INFO | train rank=0 | step=36360/100000 progress=36.4% loss=0.067780 lr=7.543784e-06 elapsed=13:25:29 avg_step=1.33s recent_step=1.18s eta=23:29:49 +2026-08-25 17:21:11 | INFO | train rank=0 | step=36370/100000 progress=36.4% loss=0.059967 lr=7.542360e-06 elapsed=13:25:41 avg_step=1.33s recent_step=1.21s eta=23:29:34 +2026-08-25 17:21:24 | INFO | train rank=0 | step=36380/100000 progress=36.4% loss=0.071106 lr=7.540936e-06 elapsed=13:25:54 avg_step=1.33s recent_step=1.31s eta=23:29:20 +2026-08-25 17:21:39 | INFO | train rank=0 | step=36390/100000 progress=36.4% loss=0.061462 lr=7.539512e-06 elapsed=13:26:09 avg_step=1.33s recent_step=1.52s eta=23:29:10 +2026-08-25 17:21:59 | INFO | train rank=0 | step=36400/100000 progress=36.4% loss=0.054169 lr=7.538088e-06 elapsed=13:26:29 avg_step=1.33s recent_step=1.99s eta=23:29:08 +2026-08-25 17:22:11 | INFO | train rank=0 | step=36410/100000 progress=36.4% loss=0.058472 lr=7.536663e-06 elapsed=13:26:41 avg_step=1.33s recent_step=1.14s eta=23:28:52 +2026-08-25 17:22:24 | INFO | train rank=0 | step=36420/100000 progress=36.4% loss=0.071930 lr=7.535238e-06 elapsed=13:26:54 avg_step=1.33s recent_step=1.33s eta=23:28:38 +2026-08-25 17:22:36 | INFO | train rank=0 | step=36430/100000 progress=36.4% loss=0.033264 lr=7.533812e-06 elapsed=13:27:06 avg_step=1.33s recent_step=1.20s eta=23:28:23 +2026-08-25 17:22:47 | INFO | train rank=0 | step=36440/100000 progress=36.4% loss=0.039680 lr=7.532387e-06 elapsed=13:27:17 avg_step=1.33s recent_step=1.16s eta=23:28:07 +2026-08-25 17:23:01 | INFO | train rank=0 | step=36450/100000 progress=36.4% loss=0.036972 lr=7.530961e-06 elapsed=13:27:30 avg_step=1.33s recent_step=1.31s eta=23:27:53 +2026-08-25 17:23:12 | INFO | train rank=0 | step=36460/100000 progress=36.5% loss=0.068726 lr=7.529535e-06 elapsed=13:27:42 avg_step=1.33s recent_step=1.18s eta=23:27:37 +2026-08-25 17:23:30 | INFO | train rank=0 | step=36470/100000 progress=36.5% loss=0.061050 lr=7.528109e-06 elapsed=13:28:00 avg_step=1.33s recent_step=1.75s eta=23:27:31 +2026-08-25 17:23:44 | INFO | train rank=0 | step=36480/100000 progress=36.5% loss=0.082306 lr=7.526682e-06 elapsed=13:28:14 avg_step=1.33s recent_step=1.42s eta=23:27:20 +2026-08-25 17:23:57 | INFO | train rank=0 | step=36490/100000 progress=36.5% loss=0.044510 lr=7.525255e-06 elapsed=13:28:26 avg_step=1.33s recent_step=1.24s eta=23:27:05 +2026-08-25 17:24:13 | INFO | train rank=0 | step=36500/100000 progress=36.5% loss=0.053375 lr=7.523828e-06 elapsed=13:28:43 avg_step=1.33s recent_step=1.67s eta=23:26:57 +2026-08-25 17:24:25 | INFO | train rank=0 | step=36510/100000 progress=36.5% loss=0.089233 lr=7.522400e-06 elapsed=13:28:55 avg_step=1.33s recent_step=1.16s eta=23:26:41 +2026-08-25 17:24:36 | INFO | train rank=0 | step=36520/100000 progress=36.5% loss=0.042542 lr=7.520972e-06 elapsed=13:29:06 avg_step=1.33s recent_step=1.15s eta=23:26:25 +2026-08-25 17:24:48 | INFO | train rank=0 | step=36530/100000 progress=36.5% loss=0.031723 lr=7.519544e-06 elapsed=13:29:18 avg_step=1.33s recent_step=1.18s eta=23:26:09 +2026-08-25 17:24:59 | INFO | train rank=0 | step=36540/100000 progress=36.5% loss=0.055817 lr=7.518116e-06 elapsed=13:29:29 avg_step=1.33s recent_step=1.12s eta=23:25:52 +2026-08-25 17:25:14 | INFO | train rank=0 | step=36550/100000 progress=36.5% loss=0.049789 lr=7.516687e-06 elapsed=13:29:44 avg_step=1.33s recent_step=1.43s eta=23:25:40 +2026-08-25 17:25:28 | INFO | train rank=0 | step=36560/100000 progress=36.6% loss=0.062401 lr=7.515259e-06 elapsed=13:29:58 avg_step=1.33s recent_step=1.46s eta=23:25:29 +2026-08-25 17:25:43 | INFO | train rank=0 | step=36570/100000 progress=36.6% loss=0.048401 lr=7.513829e-06 elapsed=13:30:13 avg_step=1.33s recent_step=1.51s eta=23:25:19 +2026-08-25 17:25:56 | INFO | train rank=0 | step=36580/100000 progress=36.6% loss=0.082108 lr=7.512400e-06 elapsed=13:30:26 avg_step=1.33s recent_step=1.26s eta=23:25:05 +2026-08-25 17:26:13 | INFO | train rank=0 | step=36590/100000 progress=36.6% loss=0.081177 lr=7.510970e-06 elapsed=13:30:42 avg_step=1.33s recent_step=1.65s eta=23:24:57 +2026-08-25 17:26:24 | INFO | train rank=0 | step=36600/100000 progress=36.6% loss=0.058624 lr=7.509540e-06 elapsed=13:30:54 avg_step=1.33s recent_step=1.15s eta=23:24:41 +2026-08-25 17:26:35 | INFO | train rank=0 | step=36610/100000 progress=36.6% loss=0.044174 lr=7.508110e-06 elapsed=13:31:05 avg_step=1.33s recent_step=1.15s eta=23:24:24 +2026-08-25 17:26:48 | INFO | train rank=0 | step=36620/100000 progress=36.6% loss=0.056366 lr=7.506679e-06 elapsed=13:31:18 avg_step=1.33s recent_step=1.25s eta=23:24:09 +2026-08-25 17:27:00 | INFO | train rank=0 | step=36630/100000 progress=36.6% loss=0.041397 lr=7.505249e-06 elapsed=13:31:30 avg_step=1.33s recent_step=1.22s eta=23:23:54 +2026-08-25 17:27:16 | INFO | train rank=0 | step=36640/100000 progress=36.6% loss=0.095947 lr=7.503818e-06 elapsed=13:31:46 avg_step=1.33s recent_step=1.61s eta=23:23:46 +2026-08-25 17:27:28 | INFO | train rank=0 | step=36650/100000 progress=36.6% loss=0.048813 lr=7.502386e-06 elapsed=13:31:57 avg_step=1.33s recent_step=1.13s eta=23:23:29 +2026-08-25 17:27:41 | INFO | train rank=0 | step=36660/100000 progress=36.7% loss=0.082428 lr=7.500955e-06 elapsed=13:32:11 avg_step=1.33s recent_step=1.37s eta=23:23:17 +2026-08-25 17:27:57 | INFO | train rank=0 | step=36670/100000 progress=36.7% loss=0.052261 lr=7.499523e-06 elapsed=13:32:27 avg_step=1.33s recent_step=1.56s eta=23:23:07 +2026-08-25 17:28:09 | INFO | train rank=0 | step=36680/100000 progress=36.7% loss=0.054291 lr=7.498090e-06 elapsed=13:32:38 avg_step=1.33s recent_step=1.17s eta=23:22:51 +2026-08-25 17:28:20 | INFO | train rank=0 | step=36690/100000 progress=36.7% loss=0.064819 lr=7.496658e-06 elapsed=13:32:50 avg_step=1.33s recent_step=1.18s eta=23:22:35 +2026-08-25 17:28:38 | INFO | train rank=0 | step=36700/100000 progress=36.7% loss=0.041290 lr=7.495225e-06 elapsed=13:33:08 avg_step=1.33s recent_step=1.81s eta=23:22:30 +2026-08-25 17:28:51 | INFO | train rank=0 | step=36710/100000 progress=36.7% loss=0.044266 lr=7.493792e-06 elapsed=13:33:21 avg_step=1.33s recent_step=1.22s eta=23:22:15 +2026-08-25 17:29:06 | INFO | train rank=0 | step=36720/100000 progress=36.7% loss=0.044144 lr=7.492359e-06 elapsed=13:33:36 avg_step=1.33s recent_step=1.53s eta=23:22:05 +2026-08-25 17:29:17 | INFO | train rank=0 | step=36730/100000 progress=36.7% loss=0.035400 lr=7.490926e-06 elapsed=13:33:47 avg_step=1.33s recent_step=1.12s eta=23:21:48 +2026-08-25 17:29:29 | INFO | train rank=0 | step=36740/100000 progress=36.7% loss=0.054886 lr=7.489492e-06 elapsed=13:33:59 avg_step=1.33s recent_step=1.15s eta=23:21:32 +2026-08-25 17:29:44 | INFO | train rank=0 | step=36750/100000 progress=36.8% loss=0.041183 lr=7.488058e-06 elapsed=13:34:14 avg_step=1.33s recent_step=1.52s eta=23:21:22 +2026-08-25 17:29:56 | INFO | train rank=0 | step=36760/100000 progress=36.8% loss=0.051697 lr=7.486623e-06 elapsed=13:34:26 avg_step=1.33s recent_step=1.19s eta=23:21:06 +2026-08-25 17:30:10 | INFO | train rank=0 | step=36770/100000 progress=36.8% loss=0.046188 lr=7.485189e-06 elapsed=13:34:40 avg_step=1.33s recent_step=1.39s eta=23:20:54 +2026-08-25 17:30:29 | INFO | train rank=0 | step=36780/100000 progress=36.8% loss=0.032410 lr=7.483754e-06 elapsed=13:34:59 avg_step=1.33s recent_step=1.91s eta=23:20:51 +2026-08-25 17:30:40 | INFO | train rank=0 | step=36790/100000 progress=36.8% loss=0.032379 lr=7.482319e-06 elapsed=13:35:10 avg_step=1.33s recent_step=1.13s eta=23:20:34 +2026-08-25 17:30:52 | INFO | train rank=0 | step=36800/100000 progress=36.8% loss=0.072937 lr=7.480883e-06 elapsed=13:35:22 avg_step=1.33s recent_step=1.21s eta=23:20:19 +2026-08-25 17:31:03 | INFO | train rank=0 | step=36810/100000 progress=36.8% loss=0.073303 lr=7.479447e-06 elapsed=13:35:33 avg_step=1.33s recent_step=1.12s eta=23:20:02 +2026-08-25 17:31:17 | INFO | train rank=0 | step=36820/100000 progress=36.8% loss=0.116455 lr=7.478011e-06 elapsed=13:35:47 avg_step=1.33s recent_step=1.35s eta=23:19:49 +2026-08-25 17:31:33 | INFO | train rank=0 | step=36830/100000 progress=36.8% loss=0.065903 lr=7.476575e-06 elapsed=13:36:03 avg_step=1.33s recent_step=1.63s eta=23:19:41 +2026-08-25 17:31:44 | INFO | train rank=0 | step=36840/100000 progress=36.8% loss=0.069031 lr=7.475139e-06 elapsed=13:36:14 avg_step=1.33s recent_step=1.12s eta=23:19:24 +2026-08-25 17:31:57 | INFO | train rank=0 | step=36850/100000 progress=36.9% loss=0.142456 lr=7.473702e-06 elapsed=13:36:26 avg_step=1.33s recent_step=1.21s eta=23:19:09 +2026-08-25 17:32:09 | INFO | train rank=0 | step=36860/100000 progress=36.9% loss=0.041794 lr=7.472265e-06 elapsed=13:36:39 avg_step=1.33s recent_step=1.23s eta=23:18:54 +2026-08-25 17:32:23 | INFO | train rank=0 | step=36870/100000 progress=36.9% loss=0.042862 lr=7.470827e-06 elapsed=13:36:53 avg_step=1.33s recent_step=1.46s eta=23:18:42 +2026-08-25 17:32:35 | INFO | train rank=0 | step=36880/100000 progress=36.9% loss=0.055740 lr=7.469390e-06 elapsed=13:37:05 avg_step=1.33s recent_step=1.19s eta=23:18:27 +2026-08-25 17:32:51 | INFO | train rank=0 | step=36890/100000 progress=36.9% loss=0.091675 lr=7.467952e-06 elapsed=13:37:21 avg_step=1.33s recent_step=1.54s eta=23:18:17 +2026-08-25 17:33:05 | INFO | train rank=0 | step=36900/100000 progress=36.9% loss=0.109253 lr=7.466514e-06 elapsed=13:37:35 avg_step=1.33s recent_step=1.44s eta=23:18:06 +2026-08-25 17:33:17 | INFO | train rank=0 | step=36910/100000 progress=36.9% loss=0.045563 lr=7.465075e-06 elapsed=13:37:47 avg_step=1.33s recent_step=1.17s eta=23:17:50 +2026-08-25 17:33:32 | INFO | train rank=0 | step=36920/100000 progress=36.9% loss=0.056854 lr=7.463637e-06 elapsed=13:38:02 avg_step=1.33s recent_step=1.49s eta=23:17:39 +2026-08-25 17:33:45 | INFO | train rank=0 | step=36930/100000 progress=36.9% loss=0.027496 lr=7.462198e-06 elapsed=13:38:15 avg_step=1.33s recent_step=1.35s eta=23:17:26 +2026-08-25 17:34:03 | INFO | train rank=0 | step=36940/100000 progress=36.9% loss=0.097717 lr=7.460759e-06 elapsed=13:38:33 avg_step=1.33s recent_step=1.80s eta=23:17:21 +2026-08-25 17:34:15 | INFO | train rank=0 | step=36950/100000 progress=37.0% loss=0.074249 lr=7.459319e-06 elapsed=13:38:45 avg_step=1.33s recent_step=1.13s eta=23:17:04 +2026-08-25 17:34:26 | INFO | train rank=0 | step=36960/100000 progress=37.0% loss=0.047684 lr=7.457879e-06 elapsed=13:38:56 avg_step=1.33s recent_step=1.17s eta=23:16:48 +2026-08-25 17:34:37 | INFO | train rank=0 | step=36970/100000 progress=37.0% loss=0.037064 lr=7.456439e-06 elapsed=13:39:07 avg_step=1.33s recent_step=1.11s eta=23:16:31 +2026-08-25 17:34:50 | INFO | train rank=0 | step=36980/100000 progress=37.0% loss=0.055878 lr=7.454999e-06 elapsed=13:39:20 avg_step=1.33s recent_step=1.24s eta=23:16:17 +2026-08-25 17:35:05 | INFO | train rank=0 | step=36990/100000 progress=37.0% loss=0.050446 lr=7.453558e-06 elapsed=13:39:35 avg_step=1.33s recent_step=1.55s eta=23:16:07 +2026-08-25 17:35:19 | INFO | train rank=0 | step=37000/100000 progress=37.0% loss=0.044380 lr=7.452118e-06 elapsed=13:39:49 avg_step=1.33s recent_step=1.35s eta=23:15:54 +2026-08-25 17:35:34 | INFO | train rank=0 | step=37010/100000 progress=37.0% loss=0.045731 lr=7.450676e-06 elapsed=13:40:04 avg_step=1.33s recent_step=1.55s eta=23:15:45 +2026-08-25 17:35:49 | INFO | train rank=0 | step=37020/100000 progress=37.0% loss=0.037766 lr=7.449235e-06 elapsed=13:40:19 avg_step=1.33s recent_step=1.43s eta=23:15:33 +2026-08-25 17:36:02 | INFO | train rank=0 | step=37030/100000 progress=37.0% loss=0.032288 lr=7.447793e-06 elapsed=13:40:32 avg_step=1.33s recent_step=1.31s eta=23:15:20 +2026-08-25 17:36:14 | INFO | train rank=0 | step=37040/100000 progress=37.0% loss=0.081848 lr=7.446351e-06 elapsed=13:40:44 avg_step=1.33s recent_step=1.22s eta=23:15:04 +2026-08-25 17:36:26 | INFO | train rank=0 | step=37050/100000 progress=37.0% loss=0.037079 lr=7.444909e-06 elapsed=13:40:56 avg_step=1.33s recent_step=1.20s eta=23:14:49 +2026-08-25 17:36:38 | INFO | train rank=0 | step=37060/100000 progress=37.1% loss=0.033745 lr=7.443467e-06 elapsed=13:41:08 avg_step=1.33s recent_step=1.20s eta=23:14:33 +2026-08-25 17:36:50 | INFO | train rank=0 | step=37070/100000 progress=37.1% loss=0.065308 lr=7.442024e-06 elapsed=13:41:20 avg_step=1.33s recent_step=1.19s eta=23:14:18 +2026-08-25 17:37:06 | INFO | train rank=0 | step=37080/100000 progress=37.1% loss=0.078491 lr=7.440581e-06 elapsed=13:41:36 avg_step=1.33s recent_step=1.57s eta=23:14:09 +2026-08-25 17:37:20 | INFO | train rank=0 | step=37090/100000 progress=37.1% loss=0.075165 lr=7.439138e-06 elapsed=13:41:50 avg_step=1.33s recent_step=1.47s eta=23:13:58 +2026-08-25 17:37:34 | INFO | train rank=0 | step=37100/100000 progress=37.1% loss=0.052704 lr=7.437694e-06 elapsed=13:42:04 avg_step=1.33s recent_step=1.34s eta=23:13:45 +2026-08-25 17:37:48 | INFO | train rank=0 | step=37110/100000 progress=37.1% loss=0.053192 lr=7.436251e-06 elapsed=13:42:18 avg_step=1.33s recent_step=1.43s eta=23:13:33 +2026-08-25 17:38:00 | INFO | train rank=0 | step=37120/100000 progress=37.1% loss=0.059616 lr=7.434807e-06 elapsed=13:42:30 avg_step=1.33s recent_step=1.20s eta=23:13:17 +2026-08-25 17:38:14 | INFO | train rank=0 | step=37130/100000 progress=37.1% loss=0.069641 lr=7.433362e-06 elapsed=13:42:44 avg_step=1.33s recent_step=1.44s eta=23:13:06 +2026-08-25 17:38:28 | INFO | train rank=0 | step=37140/100000 progress=37.1% loss=0.038025 lr=7.431918e-06 elapsed=13:42:58 avg_step=1.33s recent_step=1.33s eta=23:12:53 +2026-08-25 17:38:40 | INFO | train rank=0 | step=37150/100000 progress=37.1% loss=0.047516 lr=7.430473e-06 elapsed=13:43:09 avg_step=1.33s recent_step=1.19s eta=23:12:37 +2026-08-25 17:38:51 | INFO | train rank=0 | step=37160/100000 progress=37.2% loss=0.045746 lr=7.429028e-06 elapsed=13:43:21 avg_step=1.33s recent_step=1.18s eta=23:12:21 +2026-08-25 17:39:09 | INFO | train rank=0 | step=37170/100000 progress=37.2% loss=0.061188 lr=7.427582e-06 elapsed=13:43:39 avg_step=1.33s recent_step=1.79s eta=23:12:16 +2026-08-25 17:39:21 | INFO | train rank=0 | step=37180/100000 progress=37.2% loss=0.072052 lr=7.426137e-06 elapsed=13:43:51 avg_step=1.33s recent_step=1.15s eta=23:11:59 +2026-08-25 17:39:33 | INFO | train rank=0 | step=37190/100000 progress=37.2% loss=0.069519 lr=7.424691e-06 elapsed=13:44:03 avg_step=1.33s recent_step=1.19s eta=23:11:44 +2026-08-25 17:39:44 | INFO | train rank=0 | step=37200/100000 progress=37.2% loss=0.073975 lr=7.423245e-06 elapsed=13:44:14 avg_step=1.33s recent_step=1.16s eta=23:11:28 +2026-08-25 17:40:00 | INFO | train rank=0 | step=37210/100000 progress=37.2% loss=0.062653 lr=7.421798e-06 elapsed=13:44:30 avg_step=1.33s recent_step=1.61s eta=23:11:19 +2026-08-25 17:40:12 | INFO | train rank=0 | step=37220/100000 progress=37.2% loss=0.026329 lr=7.420352e-06 elapsed=13:44:42 avg_step=1.33s recent_step=1.14s eta=23:11:02 +2026-08-25 17:40:27 | INFO | train rank=0 | step=37230/100000 progress=37.2% loss=0.033569 lr=7.418905e-06 elapsed=13:44:57 avg_step=1.33s recent_step=1.49s eta=23:10:52 +2026-08-25 17:40:41 | INFO | train rank=0 | step=37240/100000 progress=37.2% loss=0.064850 lr=7.417457e-06 elapsed=13:45:11 avg_step=1.33s recent_step=1.48s eta=23:10:41 +2026-08-25 17:40:57 | INFO | train rank=0 | step=37250/100000 progress=37.2% loss=0.061066 lr=7.416010e-06 elapsed=13:45:26 avg_step=1.33s recent_step=1.51s eta=23:10:31 +2026-08-25 17:41:10 | INFO | train rank=0 | step=37260/100000 progress=37.3% loss=0.077454 lr=7.414562e-06 elapsed=13:45:40 avg_step=1.33s recent_step=1.33s eta=23:10:18 +2026-08-25 17:41:23 | INFO | train rank=0 | step=37270/100000 progress=37.3% loss=0.034096 lr=7.413114e-06 elapsed=13:45:53 avg_step=1.33s recent_step=1.28s eta=23:10:04 +2026-08-25 17:41:37 | INFO | train rank=0 | step=37280/100000 progress=37.3% loss=0.059113 lr=7.411666e-06 elapsed=13:46:07 avg_step=1.33s recent_step=1.39s eta=23:09:51 +2026-08-25 17:41:53 | INFO | train rank=0 | step=37290/100000 progress=37.3% loss=0.039612 lr=7.410217e-06 elapsed=13:46:23 avg_step=1.33s recent_step=1.61s eta=23:09:43 +2026-08-25 17:42:04 | INFO | train rank=0 | step=37300/100000 progress=37.3% loss=0.046906 lr=7.408768e-06 elapsed=13:46:34 avg_step=1.33s recent_step=1.15s eta=23:09:26 +2026-08-25 17:42:17 | INFO | train rank=0 | step=37310/100000 progress=37.3% loss=0.051788 lr=7.407319e-06 elapsed=13:46:47 avg_step=1.33s recent_step=1.28s eta=23:09:12 +2026-08-25 17:42:31 | INFO | train rank=0 | step=37320/100000 progress=37.3% loss=0.047913 lr=7.405870e-06 elapsed=13:47:01 avg_step=1.33s recent_step=1.37s eta=23:09:00 +2026-08-25 17:42:42 | INFO | train rank=0 | step=37330/100000 progress=37.3% loss=0.080544 lr=7.404420e-06 elapsed=13:47:12 avg_step=1.33s recent_step=1.16s eta=23:08:43 +2026-08-25 17:42:57 | INFO | train rank=0 | step=37340/100000 progress=37.3% loss=0.060547 lr=7.402970e-06 elapsed=13:47:27 avg_step=1.33s recent_step=1.46s eta=23:08:32 +2026-08-25 17:43:10 | INFO | train rank=0 | step=37350/100000 progress=37.4% loss=0.059570 lr=7.401520e-06 elapsed=13:47:40 avg_step=1.33s recent_step=1.35s eta=23:08:19 +2026-08-25 17:43:23 | INFO | train rank=0 | step=37360/100000 progress=37.4% loss=0.058960 lr=7.400070e-06 elapsed=13:47:52 avg_step=1.33s recent_step=1.22s eta=23:08:04 +2026-08-25 17:43:41 | INFO | train rank=0 | step=37370/100000 progress=37.4% loss=0.046234 lr=7.398619e-06 elapsed=13:48:11 avg_step=1.33s recent_step=1.83s eta=23:07:59 +2026-08-25 17:43:54 | INFO | train rank=0 | step=37380/100000 progress=37.4% loss=0.030258 lr=7.397168e-06 elapsed=13:48:24 avg_step=1.33s recent_step=1.31s eta=23:07:46 +2026-08-25 17:44:06 | INFO | train rank=0 | step=37390/100000 progress=37.4% loss=0.093384 lr=7.395717e-06 elapsed=13:48:36 avg_step=1.33s recent_step=1.20s eta=23:07:30 +2026-08-25 17:44:17 | INFO | train rank=0 | step=37400/100000 progress=37.4% loss=0.044601 lr=7.394266e-06 elapsed=13:48:47 avg_step=1.33s recent_step=1.14s eta=23:07:14 +2026-08-25 17:44:32 | INFO | train rank=0 | step=37410/100000 progress=37.4% loss=0.079071 lr=7.392814e-06 elapsed=13:49:02 avg_step=1.33s recent_step=1.48s eta=23:07:03 +2026-08-25 17:44:44 | INFO | train rank=0 | step=37420/100000 progress=37.4% loss=0.034698 lr=7.391362e-06 elapsed=13:49:14 avg_step=1.33s recent_step=1.17s eta=23:06:47 +2026-08-25 17:44:59 | INFO | train rank=0 | step=37430/100000 progress=37.4% loss=0.098724 lr=7.389910e-06 elapsed=13:49:29 avg_step=1.33s recent_step=1.47s eta=23:06:36 +2026-08-25 17:45:11 | INFO | train rank=0 | step=37440/100000 progress=37.4% loss=0.040314 lr=7.388457e-06 elapsed=13:49:41 avg_step=1.33s recent_step=1.24s eta=23:06:21 +2026-08-25 17:45:26 | INFO | train rank=0 | step=37450/100000 progress=37.5% loss=0.052170 lr=7.387005e-06 elapsed=13:49:56 avg_step=1.33s recent_step=1.46s eta=23:06:10 +2026-08-25 17:45:39 | INFO | train rank=0 | step=37460/100000 progress=37.5% loss=0.055099 lr=7.385552e-06 elapsed=13:50:09 avg_step=1.33s recent_step=1.34s eta=23:05:57 +2026-08-25 17:45:51 | INFO | train rank=0 | step=37470/100000 progress=37.5% loss=0.056000 lr=7.384098e-06 elapsed=13:50:21 avg_step=1.33s recent_step=1.19s eta=23:05:42 +2026-08-25 17:46:02 | INFO | train rank=0 | step=37480/100000 progress=37.5% loss=0.043892 lr=7.382645e-06 elapsed=13:50:32 avg_step=1.33s recent_step=1.13s eta=23:05:25 +2026-08-25 17:46:22 | INFO | train rank=0 | step=37490/100000 progress=37.5% loss=0.068375 lr=7.381191e-06 elapsed=13:50:52 avg_step=1.33s recent_step=1.96s eta=23:05:22 +2026-08-25 17:46:38 | INFO | train rank=0 | step=37500/100000 progress=37.5% loss=0.030495 lr=7.379737e-06 elapsed=13:51:08 avg_step=1.33s recent_step=1.59s eta=23:05:13 +2026-08-25 17:46:49 | INFO | train rank=0 | step=37510/100000 progress=37.5% loss=0.044128 lr=7.378283e-06 elapsed=13:51:19 avg_step=1.33s recent_step=1.14s eta=23:04:57 +2026-08-25 17:47:02 | INFO | train rank=0 | step=37520/100000 progress=37.5% loss=0.056015 lr=7.376828e-06 elapsed=13:51:32 avg_step=1.33s recent_step=1.26s eta=23:04:42 +2026-08-25 17:47:17 | INFO | train rank=0 | step=37530/100000 progress=37.5% loss=0.036819 lr=7.375373e-06 elapsed=13:51:47 avg_step=1.33s recent_step=1.55s eta=23:04:32 +2026-08-25 17:47:30 | INFO | train rank=0 | step=37540/100000 progress=37.5% loss=0.033722 lr=7.373918e-06 elapsed=13:52:00 avg_step=1.33s recent_step=1.29s eta=23:04:18 +2026-08-25 17:47:44 | INFO | train rank=0 | step=37550/100000 progress=37.5% loss=0.063507 lr=7.372463e-06 elapsed=13:52:14 avg_step=1.33s recent_step=1.39s eta=23:04:06 +2026-08-25 17:47:56 | INFO | train rank=0 | step=37560/100000 progress=37.6% loss=0.058685 lr=7.371007e-06 elapsed=13:52:25 avg_step=1.33s recent_step=1.16s eta=23:03:50 +2026-08-25 17:48:07 | INFO | train rank=0 | step=37570/100000 progress=37.6% loss=0.047104 lr=7.369551e-06 elapsed=13:52:37 avg_step=1.33s recent_step=1.19s eta=23:03:34 +2026-08-25 17:48:19 | INFO | train rank=0 | step=37580/100000 progress=37.6% loss=0.069687 lr=7.368095e-06 elapsed=13:52:49 avg_step=1.33s recent_step=1.17s eta=23:03:18 +2026-08-25 17:48:31 | INFO | train rank=0 | step=37590/100000 progress=37.6% loss=0.059174 lr=7.366639e-06 elapsed=13:53:01 avg_step=1.33s recent_step=1.15s eta=23:03:02 +2026-08-25 17:48:42 | INFO | train rank=0 | step=37600/100000 progress=37.6% loss=0.074738 lr=7.365182e-06 elapsed=13:53:12 avg_step=1.33s recent_step=1.16s eta=23:02:46 +2026-08-25 17:48:56 | INFO | train rank=0 | step=37610/100000 progress=37.6% loss=0.063995 lr=7.363725e-06 elapsed=13:53:26 avg_step=1.33s recent_step=1.35s eta=23:02:33 +2026-08-25 17:49:07 | INFO | train rank=0 | step=37620/100000 progress=37.6% loss=0.066284 lr=7.362268e-06 elapsed=13:53:37 avg_step=1.33s recent_step=1.14s eta=23:02:17 +2026-08-25 17:49:22 | INFO | train rank=0 | step=37630/100000 progress=37.6% loss=0.068237 lr=7.360811e-06 elapsed=13:53:52 avg_step=1.33s recent_step=1.52s eta=23:02:07 +2026-08-25 17:49:33 | INFO | train rank=0 | step=37640/100000 progress=37.6% loss=0.051132 lr=7.359353e-06 elapsed=13:54:03 avg_step=1.33s recent_step=1.10s eta=23:01:49 +2026-08-25 17:49:47 | INFO | train rank=0 | step=37650/100000 progress=37.6% loss=0.049911 lr=7.357895e-06 elapsed=13:54:17 avg_step=1.33s recent_step=1.33s eta=23:01:36 +2026-08-25 17:49:59 | INFO | train rank=0 | step=37660/100000 progress=37.7% loss=0.053925 lr=7.356437e-06 elapsed=13:54:28 avg_step=1.33s recent_step=1.18s eta=23:01:20 +2026-08-25 17:50:13 | INFO | train rank=0 | step=37670/100000 progress=37.7% loss=0.048737 lr=7.354978e-06 elapsed=13:54:43 avg_step=1.33s recent_step=1.41s eta=23:01:08 +2026-08-25 17:50:26 | INFO | train rank=0 | step=37680/100000 progress=37.7% loss=0.046440 lr=7.353520e-06 elapsed=13:54:56 avg_step=1.33s recent_step=1.38s eta=23:00:56 +2026-08-25 17:50:41 | INFO | train rank=0 | step=37690/100000 progress=37.7% loss=0.046906 lr=7.352061e-06 elapsed=13:55:11 avg_step=1.33s recent_step=1.47s eta=23:00:45 +2026-08-25 17:50:53 | INFO | train rank=0 | step=37700/100000 progress=37.7% loss=0.078979 lr=7.350601e-06 elapsed=13:55:23 avg_step=1.33s recent_step=1.19s eta=23:00:29 +2026-08-25 17:51:07 | INFO | train rank=0 | step=37710/100000 progress=37.7% loss=0.053879 lr=7.349142e-06 elapsed=13:55:37 avg_step=1.33s recent_step=1.40s eta=23:00:17 +2026-08-25 17:51:20 | INFO | train rank=0 | step=37720/100000 progress=37.7% loss=0.041977 lr=7.347682e-06 elapsed=13:55:50 avg_step=1.33s recent_step=1.33s eta=23:00:04 +2026-08-25 17:51:33 | INFO | train rank=0 | step=37730/100000 progress=37.7% loss=0.059998 lr=7.346222e-06 elapsed=13:56:03 avg_step=1.33s recent_step=1.24s eta=22:59:49 +2026-08-25 17:51:45 | INFO | train rank=0 | step=37740/100000 progress=37.7% loss=0.082123 lr=7.344762e-06 elapsed=13:56:15 avg_step=1.33s recent_step=1.26s eta=22:59:35 +2026-08-25 17:51:59 | INFO | train rank=0 | step=37750/100000 progress=37.8% loss=0.069031 lr=7.343301e-06 elapsed=13:56:29 avg_step=1.33s recent_step=1.41s eta=22:59:23 +2026-08-25 17:52:11 | INFO | train rank=0 | step=37760/100000 progress=37.8% loss=0.046700 lr=7.341841e-06 elapsed=13:56:41 avg_step=1.33s recent_step=1.20s eta=22:59:07 +2026-08-25 17:52:24 | INFO | train rank=0 | step=37770/100000 progress=37.8% loss=0.072998 lr=7.340380e-06 elapsed=13:56:54 avg_step=1.33s recent_step=1.26s eta=22:58:53 +2026-08-25 17:52:40 | INFO | train rank=0 | step=37780/100000 progress=37.8% loss=0.041794 lr=7.338918e-06 elapsed=13:57:10 avg_step=1.33s recent_step=1.58s eta=22:58:44 +2026-08-25 17:52:55 | INFO | train rank=0 | step=37790/100000 progress=37.8% loss=0.047455 lr=7.337457e-06 elapsed=13:57:25 avg_step=1.33s recent_step=1.50s eta=22:58:33 +2026-08-25 17:53:06 | INFO | train rank=0 | step=37800/100000 progress=37.8% loss=0.053909 lr=7.335995e-06 elapsed=13:57:36 avg_step=1.33s recent_step=1.15s eta=22:58:17 +2026-08-25 17:53:20 | INFO | train rank=0 | step=37810/100000 progress=37.8% loss=0.046707 lr=7.334533e-06 elapsed=13:57:50 avg_step=1.33s recent_step=1.40s eta=22:58:05 +2026-08-25 17:53:33 | INFO | train rank=0 | step=37820/100000 progress=37.8% loss=0.043533 lr=7.333071e-06 elapsed=13:58:02 avg_step=1.33s recent_step=1.22s eta=22:57:50 +2026-08-25 17:53:45 | INFO | train rank=0 | step=37830/100000 progress=37.8% loss=0.029442 lr=7.331608e-06 elapsed=13:58:15 avg_step=1.33s recent_step=1.26s eta=22:57:36 +2026-08-25 17:53:58 | INFO | train rank=0 | step=37840/100000 progress=37.8% loss=0.056274 lr=7.330145e-06 elapsed=13:58:28 avg_step=1.33s recent_step=1.24s eta=22:57:21 +2026-08-25 17:54:14 | INFO | train rank=0 | step=37850/100000 progress=37.9% loss=0.060608 lr=7.328682e-06 elapsed=13:58:44 avg_step=1.33s recent_step=1.65s eta=22:57:13 +2026-08-25 17:54:29 | INFO | train rank=0 | step=37860/100000 progress=37.9% loss=0.049728 lr=7.327219e-06 elapsed=13:58:59 avg_step=1.33s recent_step=1.45s eta=22:57:01 +2026-08-25 17:54:41 | INFO | train rank=0 | step=37870/100000 progress=37.9% loss=0.053711 lr=7.325755e-06 elapsed=13:59:11 avg_step=1.33s recent_step=1.20s eta=22:56:46 +2026-08-25 17:54:52 | INFO | train rank=0 | step=37880/100000 progress=37.9% loss=0.039062 lr=7.324292e-06 elapsed=13:59:22 avg_step=1.33s recent_step=1.17s eta=22:56:30 +2026-08-25 17:55:05 | INFO | train rank=0 | step=37890/100000 progress=37.9% loss=0.050156 lr=7.322827e-06 elapsed=13:59:35 avg_step=1.33s recent_step=1.29s eta=22:56:16 +2026-08-25 17:55:22 | INFO | train rank=0 | step=37900/100000 progress=37.9% loss=0.033882 lr=7.321363e-06 elapsed=13:59:51 avg_step=1.33s recent_step=1.64s eta=22:56:08 +2026-08-25 17:55:34 | INFO | train rank=0 | step=37910/100000 progress=37.9% loss=0.044464 lr=7.319899e-06 elapsed=14:00:04 avg_step=1.33s recent_step=1.22s eta=22:55:53 +2026-08-25 17:55:48 | INFO | train rank=0 | step=37920/100000 progress=37.9% loss=0.057190 lr=7.318434e-06 elapsed=14:00:18 avg_step=1.33s recent_step=1.41s eta=22:55:41 +2026-08-25 17:56:02 | INFO | train rank=0 | step=37930/100000 progress=37.9% loss=0.040321 lr=7.316969e-06 elapsed=14:00:31 avg_step=1.33s recent_step=1.37s eta=22:55:28 +2026-08-25 17:56:14 | INFO | train rank=0 | step=37940/100000 progress=37.9% loss=0.051376 lr=7.315503e-06 elapsed=14:00:44 avg_step=1.33s recent_step=1.21s eta=22:55:13 +2026-08-25 17:56:27 | INFO | train rank=0 | step=37950/100000 progress=38.0% loss=0.044144 lr=7.314038e-06 elapsed=14:00:57 avg_step=1.33s recent_step=1.37s eta=22:55:00 +2026-08-25 17:56:39 | INFO | train rank=0 | step=37960/100000 progress=38.0% loss=0.047035 lr=7.312572e-06 elapsed=14:01:09 avg_step=1.33s recent_step=1.21s eta=22:54:45 +2026-08-25 17:56:54 | INFO | train rank=0 | step=37970/100000 progress=38.0% loss=0.052010 lr=7.311106e-06 elapsed=14:01:23 avg_step=1.33s recent_step=1.42s eta=22:54:33 +2026-08-25 17:57:06 | INFO | train rank=0 | step=37980/100000 progress=38.0% loss=0.085114 lr=7.309639e-06 elapsed=14:01:36 avg_step=1.33s recent_step=1.20s eta=22:54:18 +2026-08-25 17:57:19 | INFO | train rank=0 | step=37990/100000 progress=38.0% loss=0.064728 lr=7.308173e-06 elapsed=14:01:48 avg_step=1.33s recent_step=1.29s eta=22:54:04 +2026-08-25 17:57:32 | INFO | train rank=0 | step=38000/100000 progress=38.0% loss=0.076813 lr=7.306706e-06 elapsed=14:02:01 avg_step=1.33s recent_step=1.30s eta=22:53:50 +2026-08-25 17:57:45 | INFO | train rank=0 | step=38010/100000 progress=38.0% loss=0.072540 lr=7.305239e-06 elapsed=14:02:15 avg_step=1.33s recent_step=1.36s eta=22:53:37 +2026-08-25 17:57:59 | INFO | train rank=0 | step=38020/100000 progress=38.0% loss=0.042404 lr=7.303771e-06 elapsed=14:02:29 avg_step=1.33s recent_step=1.40s eta=22:53:25 +2026-08-25 17:58:12 | INFO | train rank=0 | step=38030/100000 progress=38.0% loss=0.083710 lr=7.302304e-06 elapsed=14:02:42 avg_step=1.33s recent_step=1.33s eta=22:53:12 +2026-08-25 17:58:28 | INFO | train rank=0 | step=38040/100000 progress=38.0% loss=0.075653 lr=7.300836e-06 elapsed=14:02:58 avg_step=1.33s recent_step=1.55s eta=22:53:02 +2026-08-25 17:58:39 | INFO | train rank=0 | step=38050/100000 progress=38.0% loss=0.096436 lr=7.299368e-06 elapsed=14:03:09 avg_step=1.33s recent_step=1.14s eta=22:52:46 +2026-08-25 17:58:56 | INFO | train rank=0 | step=38060/100000 progress=38.1% loss=0.053864 lr=7.297899e-06 elapsed=14:03:26 avg_step=1.33s recent_step=1.67s eta=22:52:38 +2026-08-25 17:59:10 | INFO | train rank=0 | step=38070/100000 progress=38.1% loss=0.040726 lr=7.296431e-06 elapsed=14:03:40 avg_step=1.33s recent_step=1.36s eta=22:52:25 +2026-08-25 17:59:25 | INFO | train rank=0 | step=38080/100000 progress=38.1% loss=0.114136 lr=7.294962e-06 elapsed=14:03:55 avg_step=1.33s recent_step=1.54s eta=22:52:15 +2026-08-25 17:59:36 | INFO | train rank=0 | step=38090/100000 progress=38.1% loss=0.072968 lr=7.293493e-06 elapsed=14:04:06 avg_step=1.33s recent_step=1.15s eta=22:51:59 +2026-08-25 17:59:48 | INFO | train rank=0 | step=38100/100000 progress=38.1% loss=0.043015 lr=7.292023e-06 elapsed=14:04:18 avg_step=1.33s recent_step=1.18s eta=22:51:43 +2026-08-25 18:00:05 | INFO | train rank=0 | step=38110/100000 progress=38.1% loss=0.046791 lr=7.290554e-06 elapsed=14:04:35 avg_step=1.33s recent_step=1.69s eta=22:51:36 +2026-08-25 18:00:18 | INFO | train rank=0 | step=38120/100000 progress=38.1% loss=0.067688 lr=7.289084e-06 elapsed=14:04:47 avg_step=1.33s recent_step=1.24s eta=22:51:21 +2026-08-25 18:00:34 | INFO | train rank=0 | step=38130/100000 progress=38.1% loss=0.055710 lr=7.287614e-06 elapsed=14:05:03 avg_step=1.33s recent_step=1.60s eta=22:51:12 +2026-08-25 18:00:48 | INFO | train rank=0 | step=38140/100000 progress=38.1% loss=0.041031 lr=7.286143e-06 elapsed=14:05:18 avg_step=1.33s recent_step=1.41s eta=22:51:00 +2026-08-25 18:00:59 | INFO | train rank=0 | step=38150/100000 progress=38.1% loss=0.057205 lr=7.284673e-06 elapsed=14:05:29 avg_step=1.33s recent_step=1.16s eta=22:50:44 +2026-08-25 18:01:12 | INFO | train rank=0 | step=38160/100000 progress=38.2% loss=0.033073 lr=7.283202e-06 elapsed=14:05:42 avg_step=1.33s recent_step=1.32s eta=22:50:31 +2026-08-25 18:01:29 | INFO | train rank=0 | step=38170/100000 progress=38.2% loss=0.038925 lr=7.281731e-06 elapsed=14:05:59 avg_step=1.33s recent_step=1.63s eta=22:50:22 +2026-08-25 18:01:42 | INFO | train rank=0 | step=38180/100000 progress=38.2% loss=0.059540 lr=7.280259e-06 elapsed=14:06:12 avg_step=1.33s recent_step=1.37s eta=22:50:09 +2026-08-25 18:01:54 | INFO | train rank=0 | step=38190/100000 progress=38.2% loss=0.034454 lr=7.278788e-06 elapsed=14:06:24 avg_step=1.33s recent_step=1.19s eta=22:49:54 +2026-08-25 18:02:06 | INFO | train rank=0 | step=38200/100000 progress=38.2% loss=0.060989 lr=7.277316e-06 elapsed=14:06:36 avg_step=1.33s recent_step=1.21s eta=22:49:39 +2026-08-25 18:02:22 | INFO | train rank=0 | step=38210/100000 progress=38.2% loss=0.090454 lr=7.275844e-06 elapsed=14:06:52 avg_step=1.33s recent_step=1.58s eta=22:49:29 +2026-08-25 18:02:36 | INFO | train rank=0 | step=38220/100000 progress=38.2% loss=0.051758 lr=7.274371e-06 elapsed=14:07:06 avg_step=1.33s recent_step=1.41s eta=22:49:17 +2026-08-25 18:02:48 | INFO | train rank=0 | step=38230/100000 progress=38.2% loss=0.043900 lr=7.272899e-06 elapsed=14:07:17 avg_step=1.33s recent_step=1.13s eta=22:49:01 +2026-08-25 18:03:01 | INFO | train rank=0 | step=38240/100000 progress=38.2% loss=0.052505 lr=7.271426e-06 elapsed=14:07:31 avg_step=1.33s recent_step=1.38s eta=22:48:48 +2026-08-25 18:03:20 | INFO | train rank=0 | step=38250/100000 progress=38.2% loss=0.055084 lr=7.269952e-06 elapsed=14:07:50 avg_step=1.33s recent_step=1.91s eta=22:48:44 +2026-08-25 18:03:32 | INFO | train rank=0 | step=38260/100000 progress=38.3% loss=0.037254 lr=7.268479e-06 elapsed=14:08:01 avg_step=1.33s recent_step=1.11s eta=22:48:27 +2026-08-25 18:03:43 | INFO | train rank=0 | step=38270/100000 progress=38.3% loss=0.048889 lr=7.267005e-06 elapsed=14:08:13 avg_step=1.33s recent_step=1.19s eta=22:48:12 +2026-08-25 18:03:55 | INFO | train rank=0 | step=38280/100000 progress=38.3% loss=0.074768 lr=7.265532e-06 elapsed=14:08:25 avg_step=1.33s recent_step=1.16s eta=22:47:56 +2026-08-25 18:04:10 | INFO | train rank=0 | step=38290/100000 progress=38.3% loss=0.051422 lr=7.264058e-06 elapsed=14:08:40 avg_step=1.33s recent_step=1.49s eta=22:47:45 +2026-08-25 18:04:22 | INFO | train rank=0 | step=38300/100000 progress=38.3% loss=0.060699 lr=7.262583e-06 elapsed=14:08:52 avg_step=1.33s recent_step=1.17s eta=22:47:29 +2026-08-25 18:04:33 | INFO | train rank=0 | step=38310/100000 progress=38.3% loss=0.038971 lr=7.261109e-06 elapsed=14:09:03 avg_step=1.33s recent_step=1.16s eta=22:47:13 +2026-08-25 18:04:49 | INFO | train rank=0 | step=38320/100000 progress=38.3% loss=0.058006 lr=7.259634e-06 elapsed=14:09:19 avg_step=1.33s recent_step=1.59s eta=22:47:04 +2026-08-25 18:05:01 | INFO | train rank=0 | step=38330/100000 progress=38.3% loss=0.049164 lr=7.258159e-06 elapsed=14:09:31 avg_step=1.33s recent_step=1.18s eta=22:46:49 +2026-08-25 18:05:12 | INFO | train rank=0 | step=38340/100000 progress=38.3% loss=0.074371 lr=7.256683e-06 elapsed=14:09:42 avg_step=1.33s recent_step=1.14s eta=22:46:32 +2026-08-25 18:05:24 | INFO | train rank=0 | step=38350/100000 progress=38.4% loss=0.036064 lr=7.255208e-06 elapsed=14:09:54 avg_step=1.33s recent_step=1.16s eta=22:46:16 +2026-08-25 18:05:37 | INFO | train rank=0 | step=38360/100000 progress=38.4% loss=0.072052 lr=7.253732e-06 elapsed=14:10:07 avg_step=1.33s recent_step=1.29s eta=22:46:02 +2026-08-25 18:05:52 | INFO | train rank=0 | step=38370/100000 progress=38.4% loss=0.068436 lr=7.252256e-06 elapsed=14:10:21 avg_step=1.33s recent_step=1.47s eta=22:45:51 +2026-08-25 18:06:05 | INFO | train rank=0 | step=38380/100000 progress=38.4% loss=0.059296 lr=7.250779e-06 elapsed=14:10:34 avg_step=1.33s recent_step=1.30s eta=22:45:37 +2026-08-25 18:06:19 | INFO | train rank=0 | step=38390/100000 progress=38.4% loss=0.049377 lr=7.249303e-06 elapsed=14:10:49 avg_step=1.33s recent_step=1.45s eta=22:45:26 +2026-08-25 18:06:31 | INFO | train rank=0 | step=38400/100000 progress=38.4% loss=0.036453 lr=7.247826e-06 elapsed=14:11:01 avg_step=1.33s recent_step=1.24s eta=22:45:11 +2026-08-25 18:06:45 | INFO | train rank=0 | step=38410/100000 progress=38.4% loss=0.079681 lr=7.246349e-06 elapsed=14:11:15 avg_step=1.33s recent_step=1.32s eta=22:44:58 +2026-08-25 18:06:57 | INFO | train rank=0 | step=38420/100000 progress=38.4% loss=0.041504 lr=7.244871e-06 elapsed=14:11:27 avg_step=1.33s recent_step=1.27s eta=22:44:44 +2026-08-25 18:07:14 | INFO | train rank=0 | step=38430/100000 progress=38.4% loss=0.075928 lr=7.243394e-06 elapsed=14:11:44 avg_step=1.33s recent_step=1.64s eta=22:44:35 +2026-08-25 18:07:25 | INFO | train rank=0 | step=38440/100000 progress=38.4% loss=0.044609 lr=7.241916e-06 elapsed=14:11:55 avg_step=1.33s recent_step=1.14s eta=22:44:19 +2026-08-25 18:07:38 | INFO | train rank=0 | step=38450/100000 progress=38.5% loss=0.045197 lr=7.240438e-06 elapsed=14:12:08 avg_step=1.33s recent_step=1.31s eta=22:44:05 +2026-08-25 18:07:52 | INFO | train rank=0 | step=38460/100000 progress=38.5% loss=0.036118 lr=7.238960e-06 elapsed=14:12:22 avg_step=1.33s recent_step=1.39s eta=22:43:53 +2026-08-25 18:08:04 | INFO | train rank=0 | step=38470/100000 progress=38.5% loss=0.069672 lr=7.237481e-06 elapsed=14:12:33 avg_step=1.33s recent_step=1.14s eta=22:43:37 +2026-08-25 18:08:16 | INFO | train rank=0 | step=38480/100000 progress=38.5% loss=0.045532 lr=7.236002e-06 elapsed=14:12:46 avg_step=1.33s recent_step=1.26s eta=22:43:22 +2026-08-25 18:08:29 | INFO | train rank=0 | step=38490/100000 progress=38.5% loss=0.074219 lr=7.234523e-06 elapsed=14:12:59 avg_step=1.33s recent_step=1.32s eta=22:43:09 +2026-08-25 18:08:42 | INFO | train rank=0 | step=38500/100000 progress=38.5% loss=0.055069 lr=7.233044e-06 elapsed=14:13:12 avg_step=1.33s recent_step=1.27s eta=22:42:54 +2026-08-25 18:08:55 | INFO | train rank=0 | step=38510/100000 progress=38.5% loss=0.045654 lr=7.231565e-06 elapsed=14:13:25 avg_step=1.33s recent_step=1.26s eta=22:42:40 +2026-08-25 18:09:07 | INFO | train rank=0 | step=38520/100000 progress=38.5% loss=0.076065 lr=7.230085e-06 elapsed=14:13:37 avg_step=1.33s recent_step=1.20s eta=22:42:25 +2026-08-25 18:09:25 | INFO | train rank=0 | step=38530/100000 progress=38.5% loss=0.054153 lr=7.228605e-06 elapsed=14:13:55 avg_step=1.33s recent_step=1.84s eta=22:42:20 +2026-08-25 18:09:37 | INFO | train rank=0 | step=38540/100000 progress=38.5% loss=0.072876 lr=7.227124e-06 elapsed=14:14:07 avg_step=1.33s recent_step=1.22s eta=22:42:04 +2026-08-25 18:09:51 | INFO | train rank=0 | step=38550/100000 progress=38.6% loss=0.059616 lr=7.225644e-06 elapsed=14:14:21 avg_step=1.33s recent_step=1.41s eta=22:41:53 +2026-08-25 18:10:04 | INFO | train rank=0 | step=38560/100000 progress=38.6% loss=0.049866 lr=7.224163e-06 elapsed=14:14:33 avg_step=1.33s recent_step=1.21s eta=22:41:37 +2026-08-25 18:10:18 | INFO | train rank=0 | step=38570/100000 progress=38.6% loss=0.059357 lr=7.222682e-06 elapsed=14:14:47 avg_step=1.33s recent_step=1.40s eta=22:41:25 +2026-08-25 18:10:29 | INFO | train rank=0 | step=38580/100000 progress=38.6% loss=0.068298 lr=7.221201e-06 elapsed=14:14:59 avg_step=1.33s recent_step=1.14s eta=22:41:09 +2026-08-25 18:10:44 | INFO | train rank=0 | step=38590/100000 progress=38.6% loss=0.087494 lr=7.219719e-06 elapsed=14:15:14 avg_step=1.33s recent_step=1.51s eta=22:40:58 +2026-08-25 18:10:55 | INFO | train rank=0 | step=38600/100000 progress=38.6% loss=0.062531 lr=7.218238e-06 elapsed=14:15:25 avg_step=1.33s recent_step=1.12s eta=22:40:42 +2026-08-25 18:11:12 | INFO | train rank=0 | step=38610/100000 progress=38.6% loss=0.050781 lr=7.216756e-06 elapsed=14:15:42 avg_step=1.33s recent_step=1.73s eta=22:40:35 +2026-08-25 18:11:24 | INFO | train rank=0 | step=38620/100000 progress=38.6% loss=0.073425 lr=7.215274e-06 elapsed=14:15:54 avg_step=1.33s recent_step=1.19s eta=22:40:19 +2026-08-25 18:11:36 | INFO | train rank=0 | step=38630/100000 progress=38.6% loss=0.050064 lr=7.213791e-06 elapsed=14:16:05 avg_step=1.33s recent_step=1.13s eta=22:40:03 +2026-08-25 18:11:48 | INFO | train rank=0 | step=38640/100000 progress=38.6% loss=0.089661 lr=7.212308e-06 elapsed=14:16:17 avg_step=1.33s recent_step=1.20s eta=22:39:47 +2026-08-25 18:11:59 | INFO | train rank=0 | step=38650/100000 progress=38.6% loss=0.034409 lr=7.210826e-06 elapsed=14:16:29 avg_step=1.33s recent_step=1.13s eta=22:39:31 +2026-08-25 18:12:13 | INFO | train rank=0 | step=38660/100000 progress=38.7% loss=0.042007 lr=7.209342e-06 elapsed=14:16:43 avg_step=1.33s recent_step=1.44s eta=22:39:19 +2026-08-25 18:12:28 | INFO | train rank=0 | step=38670/100000 progress=38.7% loss=0.060242 lr=7.207859e-06 elapsed=14:16:58 avg_step=1.33s recent_step=1.45s eta=22:39:08 +2026-08-25 18:12:42 | INFO | train rank=0 | step=38680/100000 progress=38.7% loss=0.070007 lr=7.206375e-06 elapsed=14:17:12 avg_step=1.33s recent_step=1.42s eta=22:38:56 +2026-08-25 18:12:56 | INFO | train rank=0 | step=38690/100000 progress=38.7% loss=0.053688 lr=7.204891e-06 elapsed=14:17:25 avg_step=1.33s recent_step=1.36s eta=22:38:43 +2026-08-25 18:13:10 | INFO | train rank=0 | step=38700/100000 progress=38.7% loss=0.060394 lr=7.203407e-06 elapsed=14:17:40 avg_step=1.33s recent_step=1.42s eta=22:38:31 +2026-08-25 18:13:21 | INFO | train rank=0 | step=38710/100000 progress=38.7% loss=0.034271 lr=7.201923e-06 elapsed=14:17:51 avg_step=1.33s recent_step=1.13s eta=22:38:15 +2026-08-25 18:13:33 | INFO | train rank=0 | step=38720/100000 progress=38.7% loss=0.060028 lr=7.200438e-06 elapsed=14:18:03 avg_step=1.33s recent_step=1.21s eta=22:38:00 +2026-08-25 18:13:44 | INFO | train rank=0 | step=38730/100000 progress=38.7% loss=0.049103 lr=7.198953e-06 elapsed=14:18:14 avg_step=1.33s recent_step=1.13s eta=22:37:43 +2026-08-25 18:13:56 | INFO | train rank=0 | step=38740/100000 progress=38.7% loss=0.061783 lr=7.197468e-06 elapsed=14:18:26 avg_step=1.33s recent_step=1.15s eta=22:37:27 +2026-08-25 18:14:08 | INFO | train rank=0 | step=38750/100000 progress=38.8% loss=0.044418 lr=7.195983e-06 elapsed=14:18:38 avg_step=1.33s recent_step=1.24s eta=22:37:12 +2026-08-25 18:14:26 | INFO | train rank=0 | step=38760/100000 progress=38.8% loss=0.056915 lr=7.194497e-06 elapsed=14:18:55 avg_step=1.33s recent_step=1.73s eta=22:37:05 +2026-08-25 18:14:40 | INFO | train rank=0 | step=38770/100000 progress=38.8% loss=0.036049 lr=7.193012e-06 elapsed=14:19:10 avg_step=1.33s recent_step=1.47s eta=22:36:54 +2026-08-25 18:14:54 | INFO | train rank=0 | step=38780/100000 progress=38.8% loss=0.042526 lr=7.191525e-06 elapsed=14:19:24 avg_step=1.33s recent_step=1.36s eta=22:36:41 +2026-08-25 18:15:05 | INFO | train rank=0 | step=38790/100000 progress=38.8% loss=0.058289 lr=7.190039e-06 elapsed=14:19:35 avg_step=1.33s recent_step=1.13s eta=22:36:25 +2026-08-25 18:15:20 | INFO | train rank=0 | step=38800/100000 progress=38.8% loss=0.039337 lr=7.188553e-06 elapsed=14:19:50 avg_step=1.33s recent_step=1.45s eta=22:36:14 +2026-08-25 18:15:35 | INFO | train rank=0 | step=38810/100000 progress=38.8% loss=0.071991 lr=7.187066e-06 elapsed=14:20:05 avg_step=1.33s recent_step=1.49s eta=22:36:03 +2026-08-25 18:15:47 | INFO | train rank=0 | step=38820/100000 progress=38.8% loss=0.045319 lr=7.185579e-06 elapsed=14:20:17 avg_step=1.33s recent_step=1.25s eta=22:35:48 +2026-08-25 18:16:01 | INFO | train rank=0 | step=38830/100000 progress=38.8% loss=0.055420 lr=7.184092e-06 elapsed=14:20:31 avg_step=1.33s recent_step=1.38s eta=22:35:36 +2026-08-25 18:16:15 | INFO | train rank=0 | step=38840/100000 progress=38.8% loss=0.110657 lr=7.182604e-06 elapsed=14:20:45 avg_step=1.33s recent_step=1.43s eta=22:35:24 +2026-08-25 18:16:27 | INFO | train rank=0 | step=38850/100000 progress=38.9% loss=0.047440 lr=7.181116e-06 elapsed=14:20:57 avg_step=1.33s recent_step=1.22s eta=22:35:09 +2026-08-25 18:16:41 | INFO | train rank=0 | step=38860/100000 progress=38.9% loss=0.123047 lr=7.179628e-06 elapsed=14:21:11 avg_step=1.33s recent_step=1.36s eta=22:34:56 +2026-08-25 18:16:55 | INFO | train rank=0 | step=38870/100000 progress=38.9% loss=0.033203 lr=7.178140e-06 elapsed=14:21:25 avg_step=1.33s recent_step=1.44s eta=22:34:45 +2026-08-25 18:17:10 | INFO | train rank=0 | step=38880/100000 progress=38.9% loss=0.068817 lr=7.176652e-06 elapsed=14:21:40 avg_step=1.33s recent_step=1.45s eta=22:34:33 +2026-08-25 18:17:21 | INFO | train rank=0 | step=38890/100000 progress=38.9% loss=0.049194 lr=7.175163e-06 elapsed=14:21:51 avg_step=1.33s recent_step=1.14s eta=22:34:17 +2026-08-25 18:17:33 | INFO | train rank=0 | step=38900/100000 progress=38.9% loss=0.048767 lr=7.173674e-06 elapsed=14:22:02 avg_step=1.33s recent_step=1.12s eta=22:34:00 +2026-08-25 18:17:46 | INFO | train rank=0 | step=38910/100000 progress=38.9% loss=0.045074 lr=7.172185e-06 elapsed=14:22:16 avg_step=1.33s recent_step=1.40s eta=22:33:48 +2026-08-25 18:17:58 | INFO | train rank=0 | step=38920/100000 progress=38.9% loss=0.060791 lr=7.170696e-06 elapsed=14:22:28 avg_step=1.33s recent_step=1.14s eta=22:33:32 +2026-08-25 18:18:11 | INFO | train rank=0 | step=38930/100000 progress=38.9% loss=0.054413 lr=7.169206e-06 elapsed=14:22:41 avg_step=1.33s recent_step=1.33s eta=22:33:18 +2026-08-25 18:18:26 | INFO | train rank=0 | step=38940/100000 progress=38.9% loss=0.063919 lr=7.167716e-06 elapsed=14:22:55 avg_step=1.33s recent_step=1.45s eta=22:33:07 +2026-08-25 18:18:37 | INFO | train rank=0 | step=38950/100000 progress=39.0% loss=0.042343 lr=7.166226e-06 elapsed=14:23:07 avg_step=1.33s recent_step=1.14s eta=22:32:51 +2026-08-25 18:18:51 | INFO | train rank=0 | step=38960/100000 progress=39.0% loss=0.040039 lr=7.164736e-06 elapsed=14:23:20 avg_step=1.33s recent_step=1.36s eta=22:32:38 +2026-08-25 18:19:03 | INFO | train rank=0 | step=38970/100000 progress=39.0% loss=0.037605 lr=7.163245e-06 elapsed=14:23:33 avg_step=1.33s recent_step=1.27s eta=22:32:24 +2026-08-25 18:19:18 | INFO | train rank=0 | step=38980/100000 progress=39.0% loss=0.034355 lr=7.161754e-06 elapsed=14:23:48 avg_step=1.33s recent_step=1.44s eta=22:32:12 +2026-08-25 18:19:32 | INFO | train rank=0 | step=38990/100000 progress=39.0% loss=0.063690 lr=7.160263e-06 elapsed=14:24:02 avg_step=1.33s recent_step=1.47s eta=22:32:01 +2026-08-25 18:19:44 | INFO | train rank=0 | step=39000/100000 progress=39.0% loss=0.044189 lr=7.158772e-06 elapsed=14:24:14 avg_step=1.33s recent_step=1.14s eta=22:31:45 +2026-08-25 18:19:56 | INFO | train rank=0 | step=39010/100000 progress=39.0% loss=0.072296 lr=7.157280e-06 elapsed=14:24:26 avg_step=1.33s recent_step=1.19s eta=22:31:29 +2026-08-25 18:20:11 | INFO | train rank=0 | step=39020/100000 progress=39.0% loss=0.050552 lr=7.155788e-06 elapsed=14:24:41 avg_step=1.33s recent_step=1.53s eta=22:31:19 +2026-08-25 18:20:22 | INFO | train rank=0 | step=39030/100000 progress=39.0% loss=0.055069 lr=7.154296e-06 elapsed=14:24:52 avg_step=1.33s recent_step=1.13s eta=22:31:03 +2026-08-25 18:20:37 | INFO | train rank=0 | step=39040/100000 progress=39.0% loss=0.048553 lr=7.152804e-06 elapsed=14:25:07 avg_step=1.33s recent_step=1.50s eta=22:30:52 +2026-08-25 18:20:49 | INFO | train rank=0 | step=39050/100000 progress=39.1% loss=0.053894 lr=7.151312e-06 elapsed=14:25:19 avg_step=1.33s recent_step=1.15s eta=22:30:36 +2026-08-25 18:21:00 | INFO | train rank=0 | step=39060/100000 progress=39.1% loss=0.047272 lr=7.149819e-06 elapsed=14:25:30 avg_step=1.33s recent_step=1.16s eta=22:30:20 +2026-08-25 18:21:14 | INFO | train rank=0 | step=39070/100000 progress=39.1% loss=0.049896 lr=7.148326e-06 elapsed=14:25:43 avg_step=1.33s recent_step=1.31s eta=22:30:06 +2026-08-25 18:21:26 | INFO | train rank=0 | step=39080/100000 progress=39.1% loss=0.062408 lr=7.146833e-06 elapsed=14:25:56 avg_step=1.33s recent_step=1.29s eta=22:29:53 +2026-08-25 18:21:40 | INFO | train rank=0 | step=39090/100000 progress=39.1% loss=0.057678 lr=7.145339e-06 elapsed=14:26:10 avg_step=1.33s recent_step=1.33s eta=22:29:39 +2026-08-25 18:21:53 | INFO | train rank=0 | step=39100/100000 progress=39.1% loss=0.053925 lr=7.143846e-06 elapsed=14:26:23 avg_step=1.33s recent_step=1.36s eta=22:29:26 +2026-08-25 18:22:05 | INFO | train rank=0 | step=39110/100000 progress=39.1% loss=0.050903 lr=7.142352e-06 elapsed=14:26:35 avg_step=1.33s recent_step=1.18s eta=22:29:11 +2026-08-25 18:22:20 | INFO | train rank=0 | step=39120/100000 progress=39.1% loss=0.120972 lr=7.140858e-06 elapsed=14:26:50 avg_step=1.33s recent_step=1.53s eta=22:29:01 +2026-08-25 18:22:34 | INFO | train rank=0 | step=39130/100000 progress=39.1% loss=0.051697 lr=7.139363e-06 elapsed=14:27:04 avg_step=1.33s recent_step=1.35s eta=22:28:48 +2026-08-25 18:22:52 | INFO | train rank=0 | step=39140/100000 progress=39.1% loss=0.066833 lr=7.137869e-06 elapsed=14:27:22 avg_step=1.33s recent_step=1.78s eta=22:28:41 +2026-08-25 18:23:03 | INFO | train rank=0 | step=39150/100000 progress=39.1% loss=0.050980 lr=7.136374e-06 elapsed=14:27:33 avg_step=1.33s recent_step=1.16s eta=22:28:25 +2026-08-25 18:23:15 | INFO | train rank=0 | step=39160/100000 progress=39.2% loss=0.093445 lr=7.134879e-06 elapsed=14:27:45 avg_step=1.33s recent_step=1.16s eta=22:28:09 +2026-08-25 18:23:29 | INFO | train rank=0 | step=39170/100000 progress=39.2% loss=0.067474 lr=7.133384e-06 elapsed=14:27:59 avg_step=1.33s recent_step=1.45s eta=22:27:58 +2026-08-25 18:23:41 | INFO | train rank=0 | step=39180/100000 progress=39.2% loss=0.048798 lr=7.131888e-06 elapsed=14:28:11 avg_step=1.33s recent_step=1.22s eta=22:27:43 +2026-08-25 18:23:53 | INFO | train rank=0 | step=39190/100000 progress=39.2% loss=0.079559 lr=7.130392e-06 elapsed=14:28:23 avg_step=1.33s recent_step=1.16s eta=22:27:27 +2026-08-25 18:24:08 | INFO | train rank=0 | step=39200/100000 progress=39.2% loss=0.053543 lr=7.128896e-06 elapsed=14:28:38 avg_step=1.33s recent_step=1.46s eta=22:27:16 +2026-08-25 18:24:23 | INFO | train rank=0 | step=39210/100000 progress=39.2% loss=0.038239 lr=7.127400e-06 elapsed=14:28:53 avg_step=1.33s recent_step=1.53s eta=22:27:05 +2026-08-25 18:24:41 | INFO | train rank=0 | step=39220/100000 progress=39.2% loss=0.079437 lr=7.125904e-06 elapsed=14:29:10 avg_step=1.33s recent_step=1.77s eta=22:26:59 +2026-08-25 18:24:52 | INFO | train rank=0 | step=39230/100000 progress=39.2% loss=0.044388 lr=7.124407e-06 elapsed=14:29:22 avg_step=1.33s recent_step=1.11s eta=22:26:42 +2026-08-25 18:25:03 | INFO | train rank=0 | step=39240/100000 progress=39.2% loss=0.075195 lr=7.122910e-06 elapsed=14:29:33 avg_step=1.33s recent_step=1.12s eta=22:26:26 +2026-08-25 18:25:17 | INFO | train rank=0 | step=39250/100000 progress=39.2% loss=0.058655 lr=7.121413e-06 elapsed=14:29:47 avg_step=1.33s recent_step=1.38s eta=22:26:13 +2026-08-25 18:25:28 | INFO | train rank=0 | step=39260/100000 progress=39.3% loss=0.058594 lr=7.119916e-06 elapsed=14:29:58 avg_step=1.33s recent_step=1.13s eta=22:25:57 +2026-08-25 18:25:43 | INFO | train rank=0 | step=39270/100000 progress=39.3% loss=0.093567 lr=7.118418e-06 elapsed=14:30:13 avg_step=1.33s recent_step=1.52s eta=22:25:46 +2026-08-25 18:25:55 | INFO | train rank=0 | step=39280/100000 progress=39.3% loss=0.049835 lr=7.116920e-06 elapsed=14:30:25 avg_step=1.33s recent_step=1.17s eta=22:25:31 +2026-08-25 18:26:10 | INFO | train rank=0 | step=39290/100000 progress=39.3% loss=0.051071 lr=7.115422e-06 elapsed=14:30:40 avg_step=1.33s recent_step=1.52s eta=22:25:20 +2026-08-25 18:26:24 | INFO | train rank=0 | step=39300/100000 progress=39.3% loss=0.052261 lr=7.113924e-06 elapsed=14:30:53 avg_step=1.33s recent_step=1.34s eta=22:25:07 +2026-08-25 18:26:37 | INFO | train rank=0 | step=39310/100000 progress=39.3% loss=0.114746 lr=7.112425e-06 elapsed=14:31:07 avg_step=1.33s recent_step=1.38s eta=22:24:55 +2026-08-25 18:26:49 | INFO | train rank=0 | step=39320/100000 progress=39.3% loss=0.060196 lr=7.110926e-06 elapsed=14:31:19 avg_step=1.33s recent_step=1.17s eta=22:24:39 +2026-08-25 18:27:02 | INFO | train rank=0 | step=39330/100000 progress=39.3% loss=0.037262 lr=7.109427e-06 elapsed=14:31:32 avg_step=1.33s recent_step=1.34s eta=22:24:26 +2026-08-25 18:27:18 | INFO | train rank=0 | step=39340/100000 progress=39.3% loss=0.074677 lr=7.107928e-06 elapsed=14:31:48 avg_step=1.33s recent_step=1.58s eta=22:24:16 +2026-08-25 18:27:30 | INFO | train rank=0 | step=39350/100000 progress=39.4% loss=0.068390 lr=7.106429e-06 elapsed=14:32:00 avg_step=1.33s recent_step=1.14s eta=22:24:00 +2026-08-25 18:27:41 | INFO | train rank=0 | step=39360/100000 progress=39.4% loss=0.099182 lr=7.104929e-06 elapsed=14:32:11 avg_step=1.33s recent_step=1.15s eta=22:23:44 +2026-08-25 18:27:54 | INFO | train rank=0 | step=39370/100000 progress=39.4% loss=0.075317 lr=7.103429e-06 elapsed=14:32:24 avg_step=1.33s recent_step=1.28s eta=22:23:30 +2026-08-25 18:28:06 | INFO | train rank=0 | step=39380/100000 progress=39.4% loss=0.035828 lr=7.101929e-06 elapsed=14:32:36 avg_step=1.33s recent_step=1.23s eta=22:23:15 +2026-08-25 18:28:22 | INFO | train rank=0 | step=39390/100000 progress=39.4% loss=0.058899 lr=7.100429e-06 elapsed=14:32:52 avg_step=1.33s recent_step=1.61s eta=22:23:06 +2026-08-25 18:28:34 | INFO | train rank=0 | step=39400/100000 progress=39.4% loss=0.058167 lr=7.098928e-06 elapsed=14:33:04 avg_step=1.33s recent_step=1.19s eta=22:22:51 +2026-08-25 18:28:49 | INFO | train rank=0 | step=39410/100000 progress=39.4% loss=0.042633 lr=7.097427e-06 elapsed=14:33:19 avg_step=1.33s recent_step=1.46s eta=22:22:40 +2026-08-25 18:29:00 | INFO | train rank=0 | step=39420/100000 progress=39.4% loss=0.043442 lr=7.095926e-06 elapsed=14:33:30 avg_step=1.33s recent_step=1.12s eta=22:22:23 +2026-08-25 18:29:14 | INFO | train rank=0 | step=39430/100000 progress=39.4% loss=0.043381 lr=7.094425e-06 elapsed=14:33:44 avg_step=1.33s recent_step=1.41s eta=22:22:11 +2026-08-25 18:29:29 | INFO | train rank=0 | step=39440/100000 progress=39.4% loss=0.042679 lr=7.092923e-06 elapsed=14:33:59 avg_step=1.33s recent_step=1.49s eta=22:22:00 +2026-08-25 18:29:40 | INFO | train rank=0 | step=39450/100000 progress=39.5% loss=0.057404 lr=7.091422e-06 elapsed=14:34:10 avg_step=1.33s recent_step=1.12s eta=22:21:44 +2026-08-25 18:29:53 | INFO | train rank=0 | step=39460/100000 progress=39.5% loss=0.041779 lr=7.089920e-06 elapsed=14:34:23 avg_step=1.33s recent_step=1.30s eta=22:21:30 +2026-08-25 18:30:07 | INFO | train rank=0 | step=39470/100000 progress=39.5% loss=0.067719 lr=7.088417e-06 elapsed=14:34:37 avg_step=1.33s recent_step=1.38s eta=22:21:17 +2026-08-25 18:30:22 | INFO | train rank=0 | step=39480/100000 progress=39.5% loss=0.057953 lr=7.086915e-06 elapsed=14:34:51 avg_step=1.33s recent_step=1.44s eta=22:21:06 +2026-08-25 18:30:37 | INFO | train rank=0 | step=39490/100000 progress=39.5% loss=0.055527 lr=7.085412e-06 elapsed=14:35:07 avg_step=1.33s recent_step=1.53s eta=22:20:56 +2026-08-25 18:30:51 | INFO | train rank=0 | step=39500/100000 progress=39.5% loss=0.043266 lr=7.083909e-06 elapsed=14:35:21 avg_step=1.33s recent_step=1.40s eta=22:20:44 +2026-08-25 18:31:03 | INFO | train rank=0 | step=39510/100000 progress=39.5% loss=0.068817 lr=7.082406e-06 elapsed=14:35:33 avg_step=1.33s recent_step=1.20s eta=22:20:28 +2026-08-25 18:31:15 | INFO | train rank=0 | step=39520/100000 progress=39.5% loss=0.049515 lr=7.080903e-06 elapsed=14:35:45 avg_step=1.33s recent_step=1.19s eta=22:20:13 +2026-08-25 18:31:26 | INFO | train rank=0 | step=39530/100000 progress=39.5% loss=0.050232 lr=7.079399e-06 elapsed=14:35:56 avg_step=1.33s recent_step=1.13s eta=22:19:56 +2026-08-25 18:31:39 | INFO | train rank=0 | step=39540/100000 progress=39.5% loss=0.081268 lr=7.077895e-06 elapsed=14:36:09 avg_step=1.33s recent_step=1.33s eta=22:19:43 +2026-08-25 18:31:51 | INFO | train rank=0 | step=39550/100000 progress=39.6% loss=0.096802 lr=7.076391e-06 elapsed=14:36:21 avg_step=1.33s recent_step=1.17s eta=22:19:27 +2026-08-25 18:32:07 | INFO | train rank=0 | step=39560/100000 progress=39.6% loss=0.052536 lr=7.074887e-06 elapsed=14:36:37 avg_step=1.33s recent_step=1.59s eta=22:19:18 +2026-08-25 18:32:19 | INFO | train rank=0 | step=39570/100000 progress=39.6% loss=0.043564 lr=7.073383e-06 elapsed=14:36:49 avg_step=1.33s recent_step=1.18s eta=22:19:03 +2026-08-25 18:32:32 | INFO | train rank=0 | step=39580/100000 progress=39.6% loss=0.051544 lr=7.071878e-06 elapsed=14:37:02 avg_step=1.33s recent_step=1.34s eta=22:18:49 +2026-08-25 18:32:49 | INFO | train rank=0 | step=39590/100000 progress=39.6% loss=0.053986 lr=7.070373e-06 elapsed=14:37:19 avg_step=1.33s recent_step=1.66s eta=22:18:41 +2026-08-25 18:33:00 | INFO | train rank=0 | step=39600/100000 progress=39.6% loss=0.072601 lr=7.068868e-06 elapsed=14:37:30 avg_step=1.33s recent_step=1.12s eta=22:18:25 +2026-08-25 18:33:11 | INFO | train rank=0 | step=39610/100000 progress=39.6% loss=0.049377 lr=7.067362e-06 elapsed=14:37:41 avg_step=1.33s recent_step=1.12s eta=22:18:08 +2026-08-25 18:33:24 | INFO | train rank=0 | step=39620/100000 progress=39.6% loss=0.054535 lr=7.065857e-06 elapsed=14:37:54 avg_step=1.33s recent_step=1.30s eta=22:17:55 +2026-08-25 18:33:35 | INFO | train rank=0 | step=39630/100000 progress=39.6% loss=0.041000 lr=7.064351e-06 elapsed=14:38:05 avg_step=1.33s recent_step=1.11s eta=22:17:38 +2026-08-25 18:33:51 | INFO | train rank=0 | step=39640/100000 progress=39.6% loss=0.035339 lr=7.062845e-06 elapsed=14:38:21 avg_step=1.33s recent_step=1.57s eta=22:17:28 +2026-08-25 18:34:06 | INFO | train rank=0 | step=39650/100000 progress=39.6% loss=0.048676 lr=7.061339e-06 elapsed=14:38:36 avg_step=1.33s recent_step=1.50s eta=22:17:17 +2026-08-25 18:34:17 | INFO | train rank=0 | step=39660/100000 progress=39.7% loss=0.074249 lr=7.059832e-06 elapsed=14:38:47 avg_step=1.33s recent_step=1.13s eta=22:17:01 +2026-08-25 18:34:30 | INFO | train rank=0 | step=39670/100000 progress=39.7% loss=0.034698 lr=7.058325e-06 elapsed=14:39:00 avg_step=1.33s recent_step=1.30s eta=22:16:47 +2026-08-25 18:34:42 | INFO | train rank=0 | step=39680/100000 progress=39.7% loss=0.043915 lr=7.056818e-06 elapsed=14:39:11 avg_step=1.33s recent_step=1.11s eta=22:16:31 +2026-08-25 18:34:56 | INFO | train rank=0 | step=39690/100000 progress=39.7% loss=0.066803 lr=7.055311e-06 elapsed=14:39:26 avg_step=1.33s recent_step=1.43s eta=22:16:19 +2026-08-25 18:35:11 | INFO | train rank=0 | step=39700/100000 progress=39.7% loss=0.077271 lr=7.053804e-06 elapsed=14:39:41 avg_step=1.33s recent_step=1.57s eta=22:16:09 +2026-08-25 18:35:26 | INFO | train rank=0 | step=39710/100000 progress=39.7% loss=0.043854 lr=7.052296e-06 elapsed=14:39:56 avg_step=1.33s recent_step=1.47s eta=22:15:58 +2026-08-25 18:35:38 | INFO | train rank=0 | step=39720/100000 progress=39.7% loss=0.062546 lr=7.050788e-06 elapsed=14:40:08 avg_step=1.33s recent_step=1.15s eta=22:15:42 +2026-08-25 18:35:51 | INFO | train rank=0 | step=39730/100000 progress=39.7% loss=0.053757 lr=7.049280e-06 elapsed=14:40:21 avg_step=1.33s recent_step=1.37s eta=22:15:30 +2026-08-25 18:36:06 | INFO | train rank=0 | step=39740/100000 progress=39.7% loss=0.059143 lr=7.047772e-06 elapsed=14:40:36 avg_step=1.33s recent_step=1.44s eta=22:15:18 +2026-08-25 18:36:18 | INFO | train rank=0 | step=39750/100000 progress=39.8% loss=0.070038 lr=7.046263e-06 elapsed=14:40:48 avg_step=1.33s recent_step=1.21s eta=22:15:03 +2026-08-25 18:36:31 | INFO | train rank=0 | step=39760/100000 progress=39.8% loss=0.094604 lr=7.044754e-06 elapsed=14:41:01 avg_step=1.33s recent_step=1.36s eta=22:14:50 +2026-08-25 18:36:47 | INFO | train rank=0 | step=39770/100000 progress=39.8% loss=0.040848 lr=7.043245e-06 elapsed=14:41:17 avg_step=1.33s recent_step=1.52s eta=22:14:40 +2026-08-25 18:37:03 | INFO | train rank=0 | step=39780/100000 progress=39.8% loss=0.038284 lr=7.041736e-06 elapsed=14:41:33 avg_step=1.33s recent_step=1.66s eta=22:14:31 +2026-08-25 18:37:18 | INFO | train rank=0 | step=39790/100000 progress=39.8% loss=0.031219 lr=7.040227e-06 elapsed=14:41:48 avg_step=1.33s recent_step=1.48s eta=22:14:20 +2026-08-25 18:37:30 | INFO | train rank=0 | step=39800/100000 progress=39.8% loss=0.086853 lr=7.038717e-06 elapsed=14:42:00 avg_step=1.33s recent_step=1.16s eta=22:14:04 +2026-08-25 18:37:44 | INFO | train rank=0 | step=39810/100000 progress=39.8% loss=0.098022 lr=7.037207e-06 elapsed=14:42:14 avg_step=1.33s recent_step=1.39s eta=22:13:52 +2026-08-25 18:37:56 | INFO | train rank=0 | step=39820/100000 progress=39.8% loss=0.042084 lr=7.035697e-06 elapsed=14:42:26 avg_step=1.33s recent_step=1.23s eta=22:13:37 +2026-08-25 18:38:09 | INFO | train rank=0 | step=39830/100000 progress=39.8% loss=0.032379 lr=7.034187e-06 elapsed=14:42:38 avg_step=1.33s recent_step=1.26s eta=22:13:23 +2026-08-25 18:38:23 | INFO | train rank=0 | step=39840/100000 progress=39.8% loss=0.043739 lr=7.032676e-06 elapsed=14:42:53 avg_step=1.33s recent_step=1.43s eta=22:13:11 +2026-08-25 18:38:38 | INFO | train rank=0 | step=39850/100000 progress=39.9% loss=0.052673 lr=7.031165e-06 elapsed=14:43:08 avg_step=1.33s recent_step=1.50s eta=22:13:00 +2026-08-25 18:38:52 | INFO | train rank=0 | step=39860/100000 progress=39.9% loss=0.053223 lr=7.029654e-06 elapsed=14:43:22 avg_step=1.33s recent_step=1.40s eta=22:12:48 +2026-08-25 18:39:04 | INFO | train rank=0 | step=39870/100000 progress=39.9% loss=0.076874 lr=7.028143e-06 elapsed=14:43:34 avg_step=1.33s recent_step=1.25s eta=22:12:34 +2026-08-25 18:39:19 | INFO | train rank=0 | step=39880/100000 progress=39.9% loss=0.040482 lr=7.026632e-06 elapsed=14:43:49 avg_step=1.33s recent_step=1.47s eta=22:12:23 +2026-08-25 18:39:33 | INFO | train rank=0 | step=39890/100000 progress=39.9% loss=0.036102 lr=7.025120e-06 elapsed=14:44:03 avg_step=1.33s recent_step=1.41s eta=22:12:11 +2026-08-25 18:39:45 | INFO | train rank=0 | step=39900/100000 progress=39.9% loss=0.045700 lr=7.023608e-06 elapsed=14:44:15 avg_step=1.33s recent_step=1.18s eta=22:11:55 +2026-08-25 18:39:59 | INFO | train rank=0 | step=39910/100000 progress=39.9% loss=0.057983 lr=7.022096e-06 elapsed=14:44:28 avg_step=1.33s recent_step=1.37s eta=22:11:42 +2026-08-25 18:40:14 | INFO | train rank=0 | step=39920/100000 progress=39.9% loss=0.050812 lr=7.020584e-06 elapsed=14:44:44 avg_step=1.33s recent_step=1.54s eta=22:11:32 +2026-08-25 18:40:27 | INFO | train rank=0 | step=39930/100000 progress=39.9% loss=0.035553 lr=7.019071e-06 elapsed=14:44:57 avg_step=1.33s recent_step=1.26s eta=22:11:18 +2026-08-25 18:40:38 | INFO | train rank=0 | step=39940/100000 progress=39.9% loss=0.069855 lr=7.017558e-06 elapsed=14:45:08 avg_step=1.33s recent_step=1.16s eta=22:11:02 +2026-08-25 18:40:51 | INFO | train rank=0 | step=39950/100000 progress=40.0% loss=0.066574 lr=7.016045e-06 elapsed=14:45:21 avg_step=1.33s recent_step=1.28s eta=22:10:48 +2026-08-25 18:41:04 | INFO | train rank=0 | step=39960/100000 progress=40.0% loss=0.038574 lr=7.014532e-06 elapsed=14:45:34 avg_step=1.33s recent_step=1.32s eta=22:10:34 +2026-08-25 18:41:19 | INFO | train rank=0 | step=39970/100000 progress=40.0% loss=0.066071 lr=7.013019e-06 elapsed=14:45:49 avg_step=1.33s recent_step=1.47s eta=22:10:23 +2026-08-25 18:41:31 | INFO | train rank=0 | step=39980/100000 progress=40.0% loss=0.055084 lr=7.011505e-06 elapsed=14:46:01 avg_step=1.33s recent_step=1.18s eta=22:10:08 +2026-08-25 18:41:48 | INFO | train rank=0 | step=39990/100000 progress=40.0% loss=0.055511 lr=7.009991e-06 elapsed=14:46:18 avg_step=1.33s recent_step=1.71s eta=22:10:00 +2026-08-25 18:41:59 | INFO | train rank=0 | step=40000/100000 progress=40.0% loss=0.045013 lr=7.008477e-06 elapsed=14:46:29 avg_step=1.33s recent_step=1.15s eta=22:09:44 +2026-08-25 18:42:11 | INFO | train rank=0 | step=40010/100000 progress=40.0% loss=0.044853 lr=7.006963e-06 elapsed=14:46:41 avg_step=1.33s recent_step=1.15s eta=22:09:28 +2026-08-25 18:42:28 | INFO | train rank=0 | step=40020/100000 progress=40.0% loss=0.051315 lr=7.005448e-06 elapsed=14:46:58 avg_step=1.33s recent_step=1.72s eta=22:09:21 +2026-08-25 18:42:43 | INFO | train rank=0 | step=40030/100000 progress=40.0% loss=0.061523 lr=7.003934e-06 elapsed=14:47:12 avg_step=1.33s recent_step=1.45s eta=22:09:09 +2026-08-25 18:42:54 | INFO | train rank=0 | step=40040/100000 progress=40.0% loss=0.059479 lr=7.002419e-06 elapsed=14:47:24 avg_step=1.33s recent_step=1.15s eta=22:08:53 +2026-08-25 18:43:07 | INFO | train rank=0 | step=40050/100000 progress=40.1% loss=0.052460 lr=7.000903e-06 elapsed=14:47:37 avg_step=1.33s recent_step=1.31s eta=22:08:40 +2026-08-25 18:43:19 | INFO | train rank=0 | step=40060/100000 progress=40.1% loss=0.084442 lr=6.999388e-06 elapsed=14:47:49 avg_step=1.33s recent_step=1.18s eta=22:08:24 +2026-08-25 18:43:34 | INFO | train rank=0 | step=40070/100000 progress=40.1% loss=0.042191 lr=6.997872e-06 elapsed=14:48:04 avg_step=1.33s recent_step=1.52s eta=22:08:14 +2026-08-25 18:43:49 | INFO | train rank=0 | step=40080/100000 progress=40.1% loss=0.036377 lr=6.996356e-06 elapsed=14:48:19 avg_step=1.33s recent_step=1.49s eta=22:08:03 +2026-08-25 18:44:02 | INFO | train rank=0 | step=40090/100000 progress=40.1% loss=0.041290 lr=6.994840e-06 elapsed=14:48:32 avg_step=1.33s recent_step=1.34s eta=22:07:50 +2026-08-25 18:44:16 | INFO | train rank=0 | step=40100/100000 progress=40.1% loss=0.083923 lr=6.993324e-06 elapsed=14:48:46 avg_step=1.33s recent_step=1.34s eta=22:07:36 +2026-08-25 18:44:27 | INFO | train rank=0 | step=40110/100000 progress=40.1% loss=0.040466 lr=6.991808e-06 elapsed=14:48:57 avg_step=1.33s recent_step=1.16s eta=22:07:21 +2026-08-25 18:44:39 | INFO | train rank=0 | step=40120/100000 progress=40.1% loss=0.045223 lr=6.990291e-06 elapsed=14:49:09 avg_step=1.33s recent_step=1.18s eta=22:07:05 +2026-08-25 18:44:50 | INFO | train rank=0 | step=40130/100000 progress=40.1% loss=0.049408 lr=6.988774e-06 elapsed=14:49:20 avg_step=1.33s recent_step=1.11s eta=22:06:48 +2026-08-25 18:45:02 | INFO | train rank=0 | step=40140/100000 progress=40.1% loss=0.052277 lr=6.987257e-06 elapsed=14:49:32 avg_step=1.33s recent_step=1.16s eta=22:06:33 +2026-08-25 18:45:14 | INFO | train rank=0 | step=40150/100000 progress=40.2% loss=0.060867 lr=6.985739e-06 elapsed=14:49:44 avg_step=1.33s recent_step=1.21s eta=22:06:18 +2026-08-25 18:45:26 | INFO | train rank=0 | step=40160/100000 progress=40.2% loss=0.044022 lr=6.984222e-06 elapsed=14:49:56 avg_step=1.33s recent_step=1.19s eta=22:06:02 +2026-08-25 18:45:45 | INFO | train rank=0 | step=40170/100000 progress=40.2% loss=0.077576 lr=6.982704e-06 elapsed=14:50:15 avg_step=1.33s recent_step=1.94s eta=22:05:58 +2026-08-25 18:45:57 | INFO | train rank=0 | step=40180/100000 progress=40.2% loss=0.091766 lr=6.981186e-06 elapsed=14:50:27 avg_step=1.33s recent_step=1.14s eta=22:05:42 +2026-08-25 18:46:11 | INFO | train rank=0 | step=40190/100000 progress=40.2% loss=0.111755 lr=6.979668e-06 elapsed=14:50:41 avg_step=1.33s recent_step=1.45s eta=22:05:30 +2026-08-25 18:46:25 | INFO | train rank=0 | step=40200/100000 progress=40.2% loss=0.066269 lr=6.978149e-06 elapsed=14:50:55 avg_step=1.33s recent_step=1.41s eta=22:05:18 +2026-08-25 18:46:40 | INFO | train rank=0 | step=40210/100000 progress=40.2% loss=0.061523 lr=6.976631e-06 elapsed=14:51:10 avg_step=1.33s recent_step=1.48s eta=22:05:07 +2026-08-25 18:46:52 | INFO | train rank=0 | step=40220/100000 progress=40.2% loss=0.040527 lr=6.975112e-06 elapsed=14:51:22 avg_step=1.33s recent_step=1.17s eta=22:04:51 +2026-08-25 18:47:09 | INFO | train rank=0 | step=40230/100000 progress=40.2% loss=0.037552 lr=6.973593e-06 elapsed=14:51:39 avg_step=1.33s recent_step=1.72s eta=22:04:44 +2026-08-25 18:47:21 | INFO | train rank=0 | step=40240/100000 progress=40.2% loss=0.078796 lr=6.972073e-06 elapsed=14:51:50 avg_step=1.33s recent_step=1.16s eta=22:04:28 +2026-08-25 18:47:32 | INFO | train rank=0 | step=40250/100000 progress=40.2% loss=0.064934 lr=6.970554e-06 elapsed=14:52:02 avg_step=1.33s recent_step=1.18s eta=22:04:13 +2026-08-25 18:47:44 | INFO | train rank=0 | step=40260/100000 progress=40.3% loss=0.057068 lr=6.969034e-06 elapsed=14:52:14 avg_step=1.33s recent_step=1.15s eta=22:03:57 +2026-08-25 18:47:58 | INFO | train rank=0 | step=40270/100000 progress=40.3% loss=0.041748 lr=6.967514e-06 elapsed=14:52:28 avg_step=1.33s recent_step=1.39s eta=22:03:44 +2026-08-25 18:48:16 | INFO | train rank=0 | step=40280/100000 progress=40.3% loss=0.049484 lr=6.965994e-06 elapsed=14:52:46 avg_step=1.33s recent_step=1.81s eta=22:03:38 +2026-08-25 18:48:27 | INFO | train rank=0 | step=40290/100000 progress=40.3% loss=0.065918 lr=6.964474e-06 elapsed=14:52:57 avg_step=1.33s recent_step=1.15s eta=22:03:22 +2026-08-25 18:48:41 | INFO | train rank=0 | step=40300/100000 progress=40.3% loss=0.057068 lr=6.962953e-06 elapsed=14:53:11 avg_step=1.33s recent_step=1.37s eta=22:03:09 +2026-08-25 18:48:53 | INFO | train rank=0 | step=40310/100000 progress=40.3% loss=0.042953 lr=6.961432e-06 elapsed=14:53:23 avg_step=1.33s recent_step=1.20s eta=22:02:54 +2026-08-25 18:49:07 | INFO | train rank=0 | step=40320/100000 progress=40.3% loss=0.046631 lr=6.959911e-06 elapsed=14:53:37 avg_step=1.33s recent_step=1.41s eta=22:02:42 +2026-08-25 18:49:19 | INFO | train rank=0 | step=40330/100000 progress=40.3% loss=0.042313 lr=6.958390e-06 elapsed=14:53:49 avg_step=1.33s recent_step=1.21s eta=22:02:27 +2026-08-25 18:49:31 | INFO | train rank=0 | step=40340/100000 progress=40.3% loss=0.053009 lr=6.956868e-06 elapsed=14:54:01 avg_step=1.33s recent_step=1.16s eta=22:02:11 +2026-08-25 18:49:45 | INFO | train rank=0 | step=40350/100000 progress=40.4% loss=0.052902 lr=6.955347e-06 elapsed=14:54:15 avg_step=1.33s recent_step=1.40s eta=22:01:59 +2026-08-25 18:49:57 | INFO | train rank=0 | step=40360/100000 progress=40.4% loss=0.065125 lr=6.953825e-06 elapsed=14:54:27 avg_step=1.33s recent_step=1.20s eta=22:01:44 +2026-08-25 18:50:14 | INFO | train rank=0 | step=40370/100000 progress=40.4% loss=0.042038 lr=6.952303e-06 elapsed=14:54:44 avg_step=1.33s recent_step=1.69s eta=22:01:36 +2026-08-25 18:50:26 | INFO | train rank=0 | step=40380/100000 progress=40.4% loss=0.055466 lr=6.950780e-06 elapsed=14:54:56 avg_step=1.33s recent_step=1.20s eta=22:01:21 +2026-08-25 18:50:38 | INFO | train rank=0 | step=40390/100000 progress=40.4% loss=0.071991 lr=6.949258e-06 elapsed=14:55:08 avg_step=1.33s recent_step=1.17s eta=22:01:05 +2026-08-25 18:50:52 | INFO | train rank=0 | step=40400/100000 progress=40.4% loss=0.085754 lr=6.947735e-06 elapsed=14:55:22 avg_step=1.33s recent_step=1.46s eta=22:00:54 +2026-08-25 18:51:04 | INFO | train rank=0 | step=40410/100000 progress=40.4% loss=0.048538 lr=6.946212e-06 elapsed=14:55:34 avg_step=1.33s recent_step=1.18s eta=22:00:38 +2026-08-25 18:51:16 | INFO | train rank=0 | step=40420/100000 progress=40.4% loss=0.055618 lr=6.944689e-06 elapsed=14:55:46 avg_step=1.33s recent_step=1.20s eta=22:00:23 +2026-08-25 18:51:30 | INFO | train rank=0 | step=40430/100000 progress=40.4% loss=0.062897 lr=6.943166e-06 elapsed=14:55:59 avg_step=1.33s recent_step=1.36s eta=22:00:10 +2026-08-25 18:51:41 | INFO | train rank=0 | step=40440/100000 progress=40.4% loss=0.038544 lr=6.941642e-06 elapsed=14:56:11 avg_step=1.33s recent_step=1.16s eta=21:59:54 +2026-08-25 18:51:54 | INFO | train rank=0 | step=40450/100000 progress=40.5% loss=0.033173 lr=6.940118e-06 elapsed=14:56:24 avg_step=1.33s recent_step=1.28s eta=21:59:40 +2026-08-25 18:52:06 | INFO | train rank=0 | step=40460/100000 progress=40.5% loss=0.053528 lr=6.938594e-06 elapsed=14:56:36 avg_step=1.33s recent_step=1.21s eta=21:59:25 +2026-08-25 18:52:21 | INFO | train rank=0 | step=40470/100000 progress=40.5% loss=0.052765 lr=6.937070e-06 elapsed=14:56:51 avg_step=1.33s recent_step=1.50s eta=21:59:14 +2026-08-25 18:52:38 | INFO | train rank=0 | step=40480/100000 progress=40.5% loss=0.038315 lr=6.935545e-06 elapsed=14:57:08 avg_step=1.33s recent_step=1.73s eta=21:59:07 +2026-08-25 18:52:50 | INFO | train rank=0 | step=40490/100000 progress=40.5% loss=0.062805 lr=6.934021e-06 elapsed=14:57:20 avg_step=1.33s recent_step=1.19s eta=21:58:52 +2026-08-25 18:53:04 | INFO | train rank=0 | step=40500/100000 progress=40.5% loss=0.054855 lr=6.932496e-06 elapsed=14:57:34 avg_step=1.33s recent_step=1.35s eta=21:58:39 +2026-08-25 18:53:16 | INFO | train rank=0 | step=40510/100000 progress=40.5% loss=0.038742 lr=6.930971e-06 elapsed=14:57:45 avg_step=1.33s recent_step=1.17s eta=21:58:23 +2026-08-25 18:53:31 | INFO | train rank=0 | step=40520/100000 progress=40.5% loss=0.049896 lr=6.929445e-06 elapsed=14:58:01 avg_step=1.33s recent_step=1.56s eta=21:58:13 +2026-08-25 18:53:47 | INFO | train rank=0 | step=40530/100000 progress=40.5% loss=0.068390 lr=6.927920e-06 elapsed=14:58:17 avg_step=1.33s recent_step=1.60s eta=21:58:04 +2026-08-25 18:53:59 | INFO | train rank=0 | step=40540/100000 progress=40.5% loss=0.063232 lr=6.926394e-06 elapsed=14:58:29 avg_step=1.33s recent_step=1.16s eta=21:57:48 +2026-08-25 18:54:10 | INFO | train rank=0 | step=40550/100000 progress=40.6% loss=0.052979 lr=6.924868e-06 elapsed=14:58:40 avg_step=1.33s recent_step=1.16s eta=21:57:32 +2026-08-25 18:54:22 | INFO | train rank=0 | step=40560/100000 progress=40.6% loss=0.078491 lr=6.923342e-06 elapsed=14:58:52 avg_step=1.33s recent_step=1.16s eta=21:57:16 +2026-08-25 18:54:34 | INFO | train rank=0 | step=40570/100000 progress=40.6% loss=0.053055 lr=6.921816e-06 elapsed=14:59:04 avg_step=1.33s recent_step=1.20s eta=21:57:01 +2026-08-25 18:54:46 | INFO | train rank=0 | step=40580/100000 progress=40.6% loss=0.063995 lr=6.920289e-06 elapsed=14:59:15 avg_step=1.33s recent_step=1.15s eta=21:56:45 +2026-08-25 18:55:02 | INFO | train rank=0 | step=40590/100000 progress=40.6% loss=0.045013 lr=6.918763e-06 elapsed=14:59:32 avg_step=1.33s recent_step=1.63s eta=21:56:37 +2026-08-25 18:55:18 | INFO | train rank=0 | step=40600/100000 progress=40.6% loss=0.047409 lr=6.917236e-06 elapsed=14:59:48 avg_step=1.33s recent_step=1.58s eta=21:56:27 +2026-08-25 18:55:29 | INFO | train rank=0 | step=40610/100000 progress=40.6% loss=0.084381 lr=6.915708e-06 elapsed=14:59:59 avg_step=1.33s recent_step=1.12s eta=21:56:11 +2026-08-25 18:55:40 | INFO | train rank=0 | step=40620/100000 progress=40.6% loss=0.051346 lr=6.914181e-06 elapsed=15:00:10 avg_step=1.33s recent_step=1.16s eta=21:55:55 +2026-08-25 18:55:56 | INFO | train rank=0 | step=40630/100000 progress=40.6% loss=0.072975 lr=6.912653e-06 elapsed=15:00:25 avg_step=1.33s recent_step=1.52s eta=21:55:44 +2026-08-25 18:56:10 | INFO | train rank=0 | step=40640/100000 progress=40.6% loss=0.063629 lr=6.911126e-06 elapsed=15:00:40 avg_step=1.33s recent_step=1.48s eta=21:55:33 +2026-08-25 18:56:24 | INFO | train rank=0 | step=40650/100000 progress=40.6% loss=0.051537 lr=6.909597e-06 elapsed=15:00:54 avg_step=1.33s recent_step=1.33s eta=21:55:20 +2026-08-25 18:56:36 | INFO | train rank=0 | step=40660/100000 progress=40.7% loss=0.067871 lr=6.908069e-06 elapsed=15:01:05 avg_step=1.33s recent_step=1.19s eta=21:55:04 +2026-08-25 18:56:47 | INFO | train rank=0 | step=40670/100000 progress=40.7% loss=0.063141 lr=6.906541e-06 elapsed=15:01:17 avg_step=1.33s recent_step=1.16s eta=21:54:49 +2026-08-25 18:57:00 | INFO | train rank=0 | step=40680/100000 progress=40.7% loss=0.047012 lr=6.905012e-06 elapsed=15:01:29 avg_step=1.33s recent_step=1.24s eta=21:54:34 +2026-08-25 18:57:15 | INFO | train rank=0 | step=40690/100000 progress=40.7% loss=0.030457 lr=6.903483e-06 elapsed=15:01:45 avg_step=1.33s recent_step=1.51s eta=21:54:23 +2026-08-25 18:57:26 | INFO | train rank=0 | step=40700/100000 progress=40.7% loss=0.049576 lr=6.901954e-06 elapsed=15:01:56 avg_step=1.33s recent_step=1.17s eta=21:54:08 +2026-08-25 18:57:43 | INFO | train rank=0 | step=40710/100000 progress=40.7% loss=0.104736 lr=6.900425e-06 elapsed=15:02:13 avg_step=1.33s recent_step=1.66s eta=21:53:59 +2026-08-25 18:57:54 | INFO | train rank=0 | step=40720/100000 progress=40.7% loss=0.050919 lr=6.898895e-06 elapsed=15:02:24 avg_step=1.33s recent_step=1.12s eta=21:53:43 +2026-08-25 18:58:07 | INFO | train rank=0 | step=40730/100000 progress=40.7% loss=0.130066 lr=6.897366e-06 elapsed=15:02:37 avg_step=1.33s recent_step=1.29s eta=21:53:29 +2026-08-25 18:58:23 | INFO | train rank=0 | step=40740/100000 progress=40.7% loss=0.045563 lr=6.895836e-06 elapsed=15:02:53 avg_step=1.33s recent_step=1.62s eta=21:53:20 +2026-08-25 18:58:37 | INFO | train rank=0 | step=40750/100000 progress=40.8% loss=0.044800 lr=6.894306e-06 elapsed=15:03:07 avg_step=1.33s recent_step=1.33s eta=21:53:07 +2026-08-25 18:58:52 | INFO | train rank=0 | step=40760/100000 progress=40.8% loss=0.036774 lr=6.892775e-06 elapsed=15:03:22 avg_step=1.33s recent_step=1.57s eta=21:52:57 +2026-08-25 18:59:03 | INFO | train rank=0 | step=40770/100000 progress=40.8% loss=0.077362 lr=6.891245e-06 elapsed=15:03:33 avg_step=1.33s recent_step=1.11s eta=21:52:40 +2026-08-25 18:59:17 | INFO | train rank=0 | step=40780/100000 progress=40.8% loss=0.044998 lr=6.889714e-06 elapsed=15:03:47 avg_step=1.33s recent_step=1.41s eta=21:52:28 +2026-08-25 18:59:31 | INFO | train rank=0 | step=40790/100000 progress=40.8% loss=0.038818 lr=6.888183e-06 elapsed=15:04:01 avg_step=1.33s recent_step=1.37s eta=21:52:16 +2026-08-25 18:59:43 | INFO | train rank=0 | step=40800/100000 progress=40.8% loss=0.067780 lr=6.886652e-06 elapsed=15:04:13 avg_step=1.33s recent_step=1.21s eta=21:52:01 +2026-08-25 19:00:00 | INFO | train rank=0 | step=40810/100000 progress=40.8% loss=0.067261 lr=6.885121e-06 elapsed=15:04:29 avg_step=1.33s recent_step=1.62s eta=21:51:52 +2026-08-25 19:00:11 | INFO | train rank=0 | step=40820/100000 progress=40.8% loss=0.047928 lr=6.883589e-06 elapsed=15:04:41 avg_step=1.33s recent_step=1.16s eta=21:51:36 +2026-08-25 19:00:27 | INFO | train rank=0 | step=40830/100000 progress=40.8% loss=0.074890 lr=6.882057e-06 elapsed=15:04:57 avg_step=1.33s recent_step=1.61s eta=21:51:26 +2026-08-25 19:00:40 | INFO | train rank=0 | step=40840/100000 progress=40.8% loss=0.049911 lr=6.880526e-06 elapsed=15:05:10 avg_step=1.33s recent_step=1.27s eta=21:51:12 +2026-08-25 19:00:51 | INFO | train rank=0 | step=40850/100000 progress=40.8% loss=0.064026 lr=6.878993e-06 elapsed=15:05:21 avg_step=1.33s recent_step=1.09s eta=21:50:55 +2026-08-25 19:01:09 | INFO | train rank=0 | step=40860/100000 progress=40.9% loss=0.044113 lr=6.877461e-06 elapsed=15:05:39 avg_step=1.33s recent_step=1.80s eta=21:50:49 +2026-08-25 19:01:20 | INFO | train rank=0 | step=40870/100000 progress=40.9% loss=0.051224 lr=6.875928e-06 elapsed=15:05:50 avg_step=1.33s recent_step=1.16s eta=21:50:33 +2026-08-25 19:01:32 | INFO | train rank=0 | step=40880/100000 progress=40.9% loss=0.051285 lr=6.874396e-06 elapsed=15:06:02 avg_step=1.33s recent_step=1.16s eta=21:50:17 +2026-08-25 19:01:46 | INFO | train rank=0 | step=40890/100000 progress=40.9% loss=0.051300 lr=6.872863e-06 elapsed=15:06:16 avg_step=1.33s recent_step=1.38s eta=21:50:05 +2026-08-25 19:01:59 | INFO | train rank=0 | step=40900/100000 progress=40.9% loss=0.047089 lr=6.871329e-06 elapsed=15:06:29 avg_step=1.33s recent_step=1.30s eta=21:49:51 +2026-08-25 19:02:13 | INFO | train rank=0 | step=40910/100000 progress=40.9% loss=0.050552 lr=6.869796e-06 elapsed=15:06:43 avg_step=1.33s recent_step=1.45s eta=21:49:40 +2026-08-25 19:02:30 | INFO | train rank=0 | step=40920/100000 progress=40.9% loss=0.060059 lr=6.868262e-06 elapsed=15:07:00 avg_step=1.33s recent_step=1.68s eta=21:49:31 +2026-08-25 19:02:42 | INFO | train rank=0 | step=40930/100000 progress=40.9% loss=0.046265 lr=6.866729e-06 elapsed=15:07:12 avg_step=1.33s recent_step=1.20s eta=21:49:16 +2026-08-25 19:02:55 | INFO | train rank=0 | step=40940/100000 progress=40.9% loss=0.039139 lr=6.865195e-06 elapsed=15:07:25 avg_step=1.33s recent_step=1.33s eta=21:49:03 +2026-08-25 19:03:10 | INFO | train rank=0 | step=40950/100000 progress=40.9% loss=0.033119 lr=6.863660e-06 elapsed=15:07:40 avg_step=1.33s recent_step=1.48s eta=21:48:51 +2026-08-25 19:03:23 | INFO | train rank=0 | step=40960/100000 progress=41.0% loss=0.029041 lr=6.862126e-06 elapsed=15:07:53 avg_step=1.33s recent_step=1.33s eta=21:48:38 +2026-08-25 19:03:38 | INFO | train rank=0 | step=40970/100000 progress=41.0% loss=0.027008 lr=6.860591e-06 elapsed=15:08:08 avg_step=1.33s recent_step=1.46s eta=21:48:27 +2026-08-25 19:03:52 | INFO | train rank=0 | step=40980/100000 progress=41.0% loss=0.063431 lr=6.859057e-06 elapsed=15:08:22 avg_step=1.33s recent_step=1.45s eta=21:48:15 +2026-08-25 19:04:05 | INFO | train rank=0 | step=40990/100000 progress=41.0% loss=0.031143 lr=6.857521e-06 elapsed=15:08:35 avg_step=1.33s recent_step=1.23s eta=21:48:01 +2026-08-25 19:04:19 | INFO | train rank=0 | step=41000/100000 progress=41.0% loss=0.046677 lr=6.855986e-06 elapsed=15:08:48 avg_step=1.33s recent_step=1.38s eta=21:47:48 +2026-08-25 19:04:32 | INFO | train rank=0 | step=41010/100000 progress=41.0% loss=0.041702 lr=6.854451e-06 elapsed=15:09:02 avg_step=1.33s recent_step=1.34s eta=21:47:35 +2026-08-25 19:04:45 | INFO | train rank=0 | step=41020/100000 progress=41.0% loss=0.036911 lr=6.852915e-06 elapsed=15:09:14 avg_step=1.33s recent_step=1.26s eta=21:47:20 +2026-08-25 19:04:59 | INFO | train rank=0 | step=41030/100000 progress=41.0% loss=0.041565 lr=6.851379e-06 elapsed=15:09:29 avg_step=1.33s recent_step=1.48s eta=21:47:09 +2026-08-25 19:05:12 | INFO | train rank=0 | step=41040/100000 progress=41.0% loss=0.052994 lr=6.849843e-06 elapsed=15:09:42 avg_step=1.33s recent_step=1.29s eta=21:46:55 +2026-08-25 19:05:26 | INFO | train rank=0 | step=41050/100000 progress=41.0% loss=0.055511 lr=6.848307e-06 elapsed=15:09:56 avg_step=1.33s recent_step=1.36s eta=21:46:43 +2026-08-25 19:05:37 | INFO | train rank=0 | step=41060/100000 progress=41.1% loss=0.046890 lr=6.846771e-06 elapsed=15:10:07 avg_step=1.33s recent_step=1.13s eta=21:46:26 +2026-08-25 19:05:49 | INFO | train rank=0 | step=41070/100000 progress=41.1% loss=0.042450 lr=6.845234e-06 elapsed=15:10:19 avg_step=1.33s recent_step=1.23s eta=21:46:12 +2026-08-25 19:06:03 | INFO | train rank=0 | step=41080/100000 progress=41.1% loss=0.062485 lr=6.843697e-06 elapsed=15:10:33 avg_step=1.33s recent_step=1.32s eta=21:45:58 +2026-08-25 19:06:14 | INFO | train rank=0 | step=41090/100000 progress=41.1% loss=0.048157 lr=6.842160e-06 elapsed=15:10:43 avg_step=1.33s recent_step=1.09s eta=21:45:42 +2026-08-25 19:06:28 | INFO | train rank=0 | step=41100/100000 progress=41.1% loss=0.049149 lr=6.840623e-06 elapsed=15:10:57 avg_step=1.33s recent_step=1.40s eta=21:45:29 +2026-08-25 19:06:47 | INFO | train rank=0 | step=41110/100000 progress=41.1% loss=0.056854 lr=6.839085e-06 elapsed=15:11:17 avg_step=1.33s recent_step=1.92s eta=21:45:24 +2026-08-25 19:06:59 | INFO | train rank=0 | step=41120/100000 progress=41.1% loss=0.055878 lr=6.837548e-06 elapsed=15:11:29 avg_step=1.33s recent_step=1.23s eta=21:45:10 +2026-08-25 19:07:14 | INFO | train rank=0 | step=41130/100000 progress=41.1% loss=0.044243 lr=6.836010e-06 elapsed=15:11:44 avg_step=1.33s recent_step=1.45s eta=21:44:58 +2026-08-25 19:07:25 | INFO | train rank=0 | step=41140/100000 progress=41.1% loss=0.069641 lr=6.834472e-06 elapsed=15:11:55 avg_step=1.33s recent_step=1.12s eta=21:44:42 +2026-08-25 19:07:36 | INFO | train rank=0 | step=41150/100000 progress=41.1% loss=0.035934 lr=6.832933e-06 elapsed=15:12:06 avg_step=1.33s recent_step=1.16s eta=21:44:26 +2026-08-25 19:07:48 | INFO | train rank=0 | step=41160/100000 progress=41.2% loss=0.050797 lr=6.831395e-06 elapsed=15:12:18 avg_step=1.33s recent_step=1.18s eta=21:44:11 +2026-08-25 19:08:02 | INFO | train rank=0 | step=41170/100000 progress=41.2% loss=0.035583 lr=6.829856e-06 elapsed=15:12:32 avg_step=1.33s recent_step=1.38s eta=21:43:58 +2026-08-25 19:08:16 | INFO | train rank=0 | step=41180/100000 progress=41.2% loss=0.053391 lr=6.828317e-06 elapsed=15:12:46 avg_step=1.33s recent_step=1.39s eta=21:43:46 +2026-08-25 19:08:27 | INFO | train rank=0 | step=41190/100000 progress=41.2% loss=0.074768 lr=6.826778e-06 elapsed=15:12:57 avg_step=1.33s recent_step=1.12s eta=21:43:29 +2026-08-25 19:08:39 | INFO | train rank=0 | step=41200/100000 progress=41.2% loss=0.071655 lr=6.825239e-06 elapsed=15:13:09 avg_step=1.33s recent_step=1.15s eta=21:43:14 +2026-08-25 19:08:53 | INFO | train rank=0 | step=41210/100000 progress=41.2% loss=0.039978 lr=6.823700e-06 elapsed=15:13:23 avg_step=1.33s recent_step=1.41s eta=21:43:01 +2026-08-25 19:09:07 | INFO | train rank=0 | step=41220/100000 progress=41.2% loss=0.111572 lr=6.822160e-06 elapsed=15:13:37 avg_step=1.33s recent_step=1.44s eta=21:42:50 +2026-08-25 19:09:25 | INFO | train rank=0 | step=41230/100000 progress=41.2% loss=0.037628 lr=6.820620e-06 elapsed=15:13:55 avg_step=1.33s recent_step=1.82s eta=21:42:43 +2026-08-25 19:09:37 | INFO | train rank=0 | step=41240/100000 progress=41.2% loss=0.099731 lr=6.819080e-06 elapsed=15:14:07 avg_step=1.33s recent_step=1.14s eta=21:42:27 +2026-08-25 19:09:51 | INFO | train rank=0 | step=41250/100000 progress=41.2% loss=0.051331 lr=6.817540e-06 elapsed=15:14:21 avg_step=1.33s recent_step=1.42s eta=21:42:15 +2026-08-25 19:10:04 | INFO | train rank=0 | step=41260/100000 progress=41.3% loss=0.075684 lr=6.815999e-06 elapsed=15:14:34 avg_step=1.33s recent_step=1.34s eta=21:42:02 +2026-08-25 19:10:16 | INFO | train rank=0 | step=41270/100000 progress=41.3% loss=0.041580 lr=6.814459e-06 elapsed=15:14:46 avg_step=1.33s recent_step=1.20s eta=21:41:47 +2026-08-25 19:10:28 | INFO | train rank=0 | step=41280/100000 progress=41.3% loss=0.079742 lr=6.812918e-06 elapsed=15:14:58 avg_step=1.33s recent_step=1.13s eta=21:41:31 +2026-08-25 19:10:41 | INFO | train rank=0 | step=41290/100000 progress=41.3% loss=0.045364 lr=6.811377e-06 elapsed=15:15:11 avg_step=1.33s recent_step=1.37s eta=21:41:18 +2026-08-25 19:10:53 | INFO | train rank=0 | step=41300/100000 progress=41.3% loss=0.057358 lr=6.809836e-06 elapsed=15:15:22 avg_step=1.33s recent_step=1.13s eta=21:41:02 +2026-08-25 19:11:05 | INFO | train rank=0 | step=41310/100000 progress=41.3% loss=0.047562 lr=6.808294e-06 elapsed=15:15:34 avg_step=1.33s recent_step=1.20s eta=21:40:47 +2026-08-25 19:11:21 | INFO | train rank=0 | step=41320/100000 progress=41.3% loss=0.081970 lr=6.806752e-06 elapsed=15:15:51 avg_step=1.33s recent_step=1.60s eta=21:40:37 +2026-08-25 19:11:35 | INFO | train rank=0 | step=41330/100000 progress=41.3% loss=0.055939 lr=6.805211e-06 elapsed=15:16:05 avg_step=1.33s recent_step=1.41s eta=21:40:25 +2026-08-25 19:11:47 | INFO | train rank=0 | step=41340/100000 progress=41.3% loss=0.060272 lr=6.803669e-06 elapsed=15:16:17 avg_step=1.33s recent_step=1.26s eta=21:40:11 +2026-08-25 19:11:59 | INFO | train rank=0 | step=41350/100000 progress=41.3% loss=0.116821 lr=6.802126e-06 elapsed=15:16:29 avg_step=1.33s recent_step=1.20s eta=21:39:56 +2026-08-25 19:12:14 | INFO | train rank=0 | step=41360/100000 progress=41.4% loss=0.067245 lr=6.800584e-06 elapsed=15:16:44 avg_step=1.33s recent_step=1.46s eta=21:39:44 +2026-08-25 19:12:30 | INFO | train rank=0 | step=41370/100000 progress=41.4% loss=0.054443 lr=6.799041e-06 elapsed=15:17:00 avg_step=1.33s recent_step=1.63s eta=21:39:35 +2026-08-25 19:12:41 | INFO | train rank=0 | step=41380/100000 progress=41.4% loss=0.046356 lr=6.797498e-06 elapsed=15:17:11 avg_step=1.33s recent_step=1.13s eta=21:39:19 +2026-08-25 19:12:57 | INFO | train rank=0 | step=41390/100000 progress=41.4% loss=0.083008 lr=6.795955e-06 elapsed=15:17:27 avg_step=1.33s recent_step=1.51s eta=21:39:08 +2026-08-25 19:13:08 | INFO | train rank=0 | step=41400/100000 progress=41.4% loss=0.045944 lr=6.794412e-06 elapsed=15:17:38 avg_step=1.33s recent_step=1.18s eta=21:38:53 +2026-08-25 19:13:20 | INFO | train rank=0 | step=41410/100000 progress=41.4% loss=0.071198 lr=6.792869e-06 elapsed=15:17:50 avg_step=1.33s recent_step=1.15s eta=21:38:37 +2026-08-25 19:13:32 | INFO | train rank=0 | step=41420/100000 progress=41.4% loss=0.074249 lr=6.791325e-06 elapsed=15:18:02 avg_step=1.33s recent_step=1.23s eta=21:38:22 +2026-08-25 19:13:47 | INFO | train rank=0 | step=41430/100000 progress=41.4% loss=0.042953 lr=6.789781e-06 elapsed=15:18:17 avg_step=1.33s recent_step=1.48s eta=21:38:11 +2026-08-25 19:13:58 | INFO | train rank=0 | step=41440/100000 progress=41.4% loss=0.049088 lr=6.788237e-06 elapsed=15:18:28 avg_step=1.33s recent_step=1.12s eta=21:37:55 +2026-08-25 19:14:15 | INFO | train rank=0 | step=41450/100000 progress=41.4% loss=0.048691 lr=6.786693e-06 elapsed=15:18:45 avg_step=1.33s recent_step=1.67s eta=21:37:47 +2026-08-25 19:14:26 | INFO | train rank=0 | step=41460/100000 progress=41.5% loss=0.050507 lr=6.785149e-06 elapsed=15:18:56 avg_step=1.33s recent_step=1.11s eta=21:37:30 +2026-08-25 19:14:38 | INFO | train rank=0 | step=41470/100000 progress=41.5% loss=0.027588 lr=6.783604e-06 elapsed=15:19:08 avg_step=1.33s recent_step=1.19s eta=21:37:15 +2026-08-25 19:14:49 | INFO | train rank=0 | step=41480/100000 progress=41.5% loss=0.031090 lr=6.782059e-06 elapsed=15:19:19 avg_step=1.33s recent_step=1.15s eta=21:36:59 +2026-08-25 19:15:01 | INFO | train rank=0 | step=41490/100000 progress=41.5% loss=0.058167 lr=6.780514e-06 elapsed=15:19:31 avg_step=1.33s recent_step=1.13s eta=21:36:43 +2026-08-25 19:15:15 | INFO | train rank=0 | step=41500/100000 progress=41.5% loss=0.037064 lr=6.778969e-06 elapsed=15:19:45 avg_step=1.33s recent_step=1.42s eta=21:36:31 +2026-08-25 19:15:27 | INFO | train rank=0 | step=41510/100000 progress=41.5% loss=0.066925 lr=6.777424e-06 elapsed=15:19:57 avg_step=1.33s recent_step=1.20s eta=21:36:16 +2026-08-25 19:15:39 | INFO | train rank=0 | step=41520/100000 progress=41.5% loss=0.045776 lr=6.775878e-06 elapsed=15:20:09 avg_step=1.33s recent_step=1.17s eta=21:36:00 +2026-08-25 19:15:51 | INFO | train rank=0 | step=41530/100000 progress=41.5% loss=0.072784 lr=6.774333e-06 elapsed=15:20:20 avg_step=1.33s recent_step=1.19s eta=21:35:45 +2026-08-25 19:16:07 | INFO | train rank=0 | step=41540/100000 progress=41.5% loss=0.057739 lr=6.772787e-06 elapsed=15:20:36 avg_step=1.33s recent_step=1.59s eta=21:35:35 +2026-08-25 19:16:19 | INFO | train rank=0 | step=41550/100000 progress=41.5% loss=0.066406 lr=6.771240e-06 elapsed=15:20:49 avg_step=1.33s recent_step=1.24s eta=21:35:21 +2026-08-25 19:16:34 | INFO | train rank=0 | step=41560/100000 progress=41.6% loss=0.046432 lr=6.769694e-06 elapsed=15:21:04 avg_step=1.33s recent_step=1.50s eta=21:35:10 +2026-08-25 19:16:45 | INFO | train rank=0 | step=41570/100000 progress=41.6% loss=0.058472 lr=6.768148e-06 elapsed=15:21:15 avg_step=1.33s recent_step=1.11s eta=21:34:54 +2026-08-25 19:17:03 | INFO | train rank=0 | step=41580/100000 progress=41.6% loss=0.046265 lr=6.766601e-06 elapsed=15:21:33 avg_step=1.33s recent_step=1.80s eta=21:34:47 +2026-08-25 19:17:17 | INFO | train rank=0 | step=41590/100000 progress=41.6% loss=0.046585 lr=6.765054e-06 elapsed=15:21:47 avg_step=1.33s recent_step=1.40s eta=21:34:35 +2026-08-25 19:17:29 | INFO | train rank=0 | step=41600/100000 progress=41.6% loss=0.082642 lr=6.763507e-06 elapsed=15:21:58 avg_step=1.33s recent_step=1.15s eta=21:34:19 +2026-08-25 19:17:40 | INFO | train rank=0 | step=41610/100000 progress=41.6% loss=0.039124 lr=6.761959e-06 elapsed=15:22:10 avg_step=1.33s recent_step=1.18s eta=21:34:03 +2026-08-25 19:17:52 | INFO | train rank=0 | step=41620/100000 progress=41.6% loss=0.045273 lr=6.760412e-06 elapsed=15:22:22 avg_step=1.33s recent_step=1.17s eta=21:33:48 +2026-08-25 19:18:06 | INFO | train rank=0 | step=41630/100000 progress=41.6% loss=0.052200 lr=6.758864e-06 elapsed=15:22:36 avg_step=1.33s recent_step=1.39s eta=21:33:35 +2026-08-25 19:18:21 | INFO | train rank=0 | step=41640/100000 progress=41.6% loss=0.046829 lr=6.757316e-06 elapsed=15:22:51 avg_step=1.33s recent_step=1.54s eta=21:33:25 +2026-08-25 19:18:35 | INFO | train rank=0 | step=41650/100000 progress=41.6% loss=0.072266 lr=6.755768e-06 elapsed=15:23:05 avg_step=1.33s recent_step=1.39s eta=21:33:13 +2026-08-25 19:18:50 | INFO | train rank=0 | step=41660/100000 progress=41.7% loss=0.048264 lr=6.754220e-06 elapsed=15:23:20 avg_step=1.33s recent_step=1.47s eta=21:33:01 +2026-08-25 19:19:01 | INFO | train rank=0 | step=41670/100000 progress=41.7% loss=0.056274 lr=6.752672e-06 elapsed=15:23:31 avg_step=1.33s recent_step=1.15s eta=21:32:45 +2026-08-25 19:19:15 | INFO | train rank=0 | step=41680/100000 progress=41.7% loss=0.033752 lr=6.751123e-06 elapsed=15:23:45 avg_step=1.33s recent_step=1.37s eta=21:32:33 +2026-08-25 19:19:32 | INFO | train rank=0 | step=41690/100000 progress=41.7% loss=0.060822 lr=6.749574e-06 elapsed=15:24:02 avg_step=1.33s recent_step=1.68s eta=21:32:24 +2026-08-25 19:19:47 | INFO | train rank=0 | step=41700/100000 progress=41.7% loss=0.089386 lr=6.748025e-06 elapsed=15:24:17 avg_step=1.33s recent_step=1.47s eta=21:32:13 +2026-08-25 19:19:59 | INFO | train rank=0 | step=41710/100000 progress=41.7% loss=0.063110 lr=6.746476e-06 elapsed=15:24:28 avg_step=1.33s recent_step=1.19s eta=21:31:58 +2026-08-25 19:20:12 | INFO | train rank=0 | step=41720/100000 progress=41.7% loss=0.061180 lr=6.744926e-06 elapsed=15:24:42 avg_step=1.33s recent_step=1.32s eta=21:31:44 +2026-08-25 19:20:28 | INFO | train rank=0 | step=41730/100000 progress=41.7% loss=0.162292 lr=6.743377e-06 elapsed=15:24:58 avg_step=1.33s recent_step=1.67s eta=21:31:36 +2026-08-25 19:20:40 | INFO | train rank=0 | step=41740/100000 progress=41.7% loss=0.057373 lr=6.741827e-06 elapsed=15:25:10 avg_step=1.33s recent_step=1.20s eta=21:31:21 +2026-08-25 19:20:55 | INFO | train rank=0 | step=41750/100000 progress=41.8% loss=0.097778 lr=6.740277e-06 elapsed=15:25:25 avg_step=1.33s recent_step=1.45s eta=21:31:09 +2026-08-25 19:21:07 | INFO | train rank=0 | step=41760/100000 progress=41.8% loss=0.034882 lr=6.738727e-06 elapsed=15:25:37 avg_step=1.33s recent_step=1.17s eta=21:30:54 +2026-08-25 19:21:22 | INFO | train rank=0 | step=41770/100000 progress=41.8% loss=0.098511 lr=6.737176e-06 elapsed=15:25:52 avg_step=1.33s recent_step=1.56s eta=21:30:44 +2026-08-25 19:21:35 | INFO | train rank=0 | step=41780/100000 progress=41.8% loss=0.028053 lr=6.735626e-06 elapsed=15:26:05 avg_step=1.33s recent_step=1.30s eta=21:30:30 +2026-08-25 19:21:50 | INFO | train rank=0 | step=41790/100000 progress=41.8% loss=0.083099 lr=6.734075e-06 elapsed=15:26:20 avg_step=1.33s recent_step=1.45s eta=21:30:18 +2026-08-25 19:22:01 | INFO | train rank=0 | step=41800/100000 progress=41.8% loss=0.055176 lr=6.732524e-06 elapsed=15:26:31 avg_step=1.33s recent_step=1.15s eta=21:30:02 +2026-08-25 19:22:17 | INFO | train rank=0 | step=41810/100000 progress=41.8% loss=0.094910 lr=6.730973e-06 elapsed=15:26:47 avg_step=1.33s recent_step=1.60s eta=21:29:53 +2026-08-25 19:22:31 | INFO | train rank=0 | step=41820/100000 progress=41.8% loss=0.060989 lr=6.729422e-06 elapsed=15:27:01 avg_step=1.33s recent_step=1.33s eta=21:29:40 +2026-08-25 19:22:42 | INFO | train rank=0 | step=41830/100000 progress=41.8% loss=0.069031 lr=6.727870e-06 elapsed=15:27:12 avg_step=1.33s recent_step=1.16s eta=21:29:24 +2026-08-25 19:22:54 | INFO | train rank=0 | step=41840/100000 progress=41.8% loss=0.066559 lr=6.726319e-06 elapsed=15:27:24 avg_step=1.33s recent_step=1.18s eta=21:29:08 +2026-08-25 19:23:08 | INFO | train rank=0 | step=41850/100000 progress=41.9% loss=0.061630 lr=6.724767e-06 elapsed=15:27:38 avg_step=1.33s recent_step=1.37s eta=21:28:56 +2026-08-25 19:23:19 | INFO | train rank=0 | step=41860/100000 progress=41.9% loss=0.048981 lr=6.723215e-06 elapsed=15:27:49 avg_step=1.33s recent_step=1.15s eta=21:28:40 +2026-08-25 19:23:32 | INFO | train rank=0 | step=41870/100000 progress=41.9% loss=0.036621 lr=6.721662e-06 elapsed=15:28:01 avg_step=1.33s recent_step=1.23s eta=21:28:25 +2026-08-25 19:23:45 | INFO | train rank=0 | step=41880/100000 progress=41.9% loss=0.063507 lr=6.720110e-06 elapsed=15:28:14 avg_step=1.33s recent_step=1.30s eta=21:28:12 +2026-08-25 19:23:59 | INFO | train rank=0 | step=41890/100000 progress=41.9% loss=0.084381 lr=6.718557e-06 elapsed=15:28:29 avg_step=1.33s recent_step=1.41s eta=21:27:59 +2026-08-25 19:24:13 | INFO | train rank=0 | step=41900/100000 progress=41.9% loss=0.093506 lr=6.717004e-06 elapsed=15:28:42 avg_step=1.33s recent_step=1.38s eta=21:27:47 +2026-08-25 19:24:27 | INFO | train rank=0 | step=41910/100000 progress=41.9% loss=0.046890 lr=6.715451e-06 elapsed=15:28:57 avg_step=1.33s recent_step=1.47s eta=21:27:36 +2026-08-25 19:24:42 | INFO | train rank=0 | step=41920/100000 progress=41.9% loss=0.037613 lr=6.713898e-06 elapsed=15:29:12 avg_step=1.33s recent_step=1.50s eta=21:27:25 +2026-08-25 19:24:54 | INFO | train rank=0 | step=41930/100000 progress=41.9% loss=0.055908 lr=6.712345e-06 elapsed=15:29:24 avg_step=1.33s recent_step=1.14s eta=21:27:09 +2026-08-25 19:25:07 | INFO | train rank=0 | step=41940/100000 progress=41.9% loss=0.035126 lr=6.710791e-06 elapsed=15:29:37 avg_step=1.33s recent_step=1.34s eta=21:26:55 +2026-08-25 19:25:20 | INFO | train rank=0 | step=41950/100000 progress=41.9% loss=0.042084 lr=6.709237e-06 elapsed=15:29:50 avg_step=1.33s recent_step=1.30s eta=21:26:42 +2026-08-25 19:25:38 | INFO | train rank=0 | step=41960/100000 progress=42.0% loss=0.103455 lr=6.707684e-06 elapsed=15:30:08 avg_step=1.33s recent_step=1.80s eta=21:26:35 +2026-08-25 19:25:49 | INFO | train rank=0 | step=41970/100000 progress=42.0% loss=0.032120 lr=6.706129e-06 elapsed=15:30:19 avg_step=1.33s recent_step=1.13s eta=21:26:19 +2026-08-25 19:26:02 | INFO | train rank=0 | step=41980/100000 progress=42.0% loss=0.089447 lr=6.704575e-06 elapsed=15:30:32 avg_step=1.33s recent_step=1.30s eta=21:26:05 +2026-08-25 19:26:14 | INFO | train rank=0 | step=41990/100000 progress=42.0% loss=0.045479 lr=6.703021e-06 elapsed=15:30:44 avg_step=1.33s recent_step=1.15s eta=21:25:49 +2026-08-25 19:26:26 | INFO | train rank=0 | step=42000/100000 progress=42.0% loss=0.068542 lr=6.701466e-06 elapsed=15:30:56 avg_step=1.33s recent_step=1.20s eta=21:25:34 +2026-08-25 19:26:39 | INFO | train rank=0 | step=42010/100000 progress=42.0% loss=0.066620 lr=6.699911e-06 elapsed=15:31:09 avg_step=1.33s recent_step=1.32s eta=21:25:21 +2026-08-25 19:26:54 | INFO | train rank=0 | step=42020/100000 progress=42.0% loss=0.065643 lr=6.698356e-06 elapsed=15:31:24 avg_step=1.33s recent_step=1.48s eta=21:25:10 +2026-08-25 19:27:07 | INFO | train rank=0 | step=42030/100000 progress=42.0% loss=0.046265 lr=6.696801e-06 elapsed=15:31:37 avg_step=1.33s recent_step=1.35s eta=21:24:57 +2026-08-25 19:27:19 | INFO | train rank=0 | step=42040/100000 progress=42.0% loss=0.060242 lr=6.695245e-06 elapsed=15:31:49 avg_step=1.33s recent_step=1.17s eta=21:24:41 +2026-08-25 19:27:34 | INFO | train rank=0 | step=42050/100000 progress=42.0% loss=0.045914 lr=6.693690e-06 elapsed=15:32:04 avg_step=1.33s recent_step=1.48s eta=21:24:30 +2026-08-25 19:27:50 | INFO | train rank=0 | step=42060/100000 progress=42.1% loss=0.063431 lr=6.692134e-06 elapsed=15:32:19 avg_step=1.33s recent_step=1.57s eta=21:24:20 +2026-08-25 19:28:01 | INFO | train rank=0 | step=42070/100000 progress=42.1% loss=0.054733 lr=6.690578e-06 elapsed=15:32:31 avg_step=1.33s recent_step=1.16s eta=21:24:04 +2026-08-25 19:28:15 | INFO | train rank=0 | step=42080/100000 progress=42.1% loss=0.080627 lr=6.689022e-06 elapsed=15:32:45 avg_step=1.33s recent_step=1.41s eta=21:23:52 +2026-08-25 19:28:27 | INFO | train rank=0 | step=42090/100000 progress=42.1% loss=0.047150 lr=6.687465e-06 elapsed=15:32:57 avg_step=1.33s recent_step=1.21s eta=21:23:37 +2026-08-25 19:28:42 | INFO | train rank=0 | step=42100/100000 progress=42.1% loss=0.057007 lr=6.685909e-06 elapsed=15:33:12 avg_step=1.33s recent_step=1.50s eta=21:23:26 +2026-08-25 19:28:55 | INFO | train rank=0 | step=42110/100000 progress=42.1% loss=0.068237 lr=6.684352e-06 elapsed=15:33:25 avg_step=1.33s recent_step=1.29s eta=21:23:12 +2026-08-25 19:29:07 | INFO | train rank=0 | step=42120/100000 progress=42.1% loss=0.067474 lr=6.682795e-06 elapsed=15:33:37 avg_step=1.33s recent_step=1.16s eta=21:22:56 +2026-08-25 19:29:21 | INFO | train rank=0 | step=42130/100000 progress=42.1% loss=0.059319 lr=6.681238e-06 elapsed=15:33:51 avg_step=1.33s recent_step=1.47s eta=21:22:45 +2026-08-25 19:29:36 | INFO | train rank=0 | step=42140/100000 progress=42.1% loss=0.044037 lr=6.679681e-06 elapsed=15:34:06 avg_step=1.33s recent_step=1.48s eta=21:22:34 +2026-08-25 19:29:50 | INFO | train rank=0 | step=42150/100000 progress=42.1% loss=0.024643 lr=6.678123e-06 elapsed=15:34:20 avg_step=1.33s recent_step=1.42s eta=21:22:22 +2026-08-25 19:30:04 | INFO | train rank=0 | step=42160/100000 progress=42.2% loss=0.024101 lr=6.676566e-06 elapsed=15:34:34 avg_step=1.33s recent_step=1.37s eta=21:22:09 +2026-08-25 19:30:16 | INFO | train rank=0 | step=42170/100000 progress=42.2% loss=0.047165 lr=6.675008e-06 elapsed=15:34:46 avg_step=1.33s recent_step=1.16s eta=21:21:53 +2026-08-25 19:30:30 | INFO | train rank=0 | step=42180/100000 progress=42.2% loss=0.027565 lr=6.673450e-06 elapsed=15:35:00 avg_step=1.33s recent_step=1.45s eta=21:21:42 +2026-08-25 19:30:42 | INFO | train rank=0 | step=42190/100000 progress=42.2% loss=0.049332 lr=6.671891e-06 elapsed=15:35:12 avg_step=1.33s recent_step=1.15s eta=21:21:26 +2026-08-25 19:30:53 | INFO | train rank=0 | step=42200/100000 progress=42.2% loss=0.056549 lr=6.670333e-06 elapsed=15:35:23 avg_step=1.33s recent_step=1.14s eta=21:21:10 +2026-08-25 19:31:10 | INFO | train rank=0 | step=42210/100000 progress=42.2% loss=0.045059 lr=6.668775e-06 elapsed=15:35:40 avg_step=1.33s recent_step=1.66s eta=21:21:01 +2026-08-25 19:31:22 | INFO | train rank=0 | step=42220/100000 progress=42.2% loss=0.064789 lr=6.667216e-06 elapsed=15:35:51 avg_step=1.33s recent_step=1.18s eta=21:20:46 +2026-08-25 19:31:33 | INFO | train rank=0 | step=42230/100000 progress=42.2% loss=0.044785 lr=6.665657e-06 elapsed=15:36:03 avg_step=1.33s recent_step=1.17s eta=21:20:30 +2026-08-25 19:31:45 | INFO | train rank=0 | step=42240/100000 progress=42.2% loss=0.043365 lr=6.664098e-06 elapsed=15:36:15 avg_step=1.33s recent_step=1.18s eta=21:20:15 +2026-08-25 19:32:00 | INFO | train rank=0 | step=42250/100000 progress=42.2% loss=0.039581 lr=6.662538e-06 elapsed=15:36:30 avg_step=1.33s recent_step=1.53s eta=21:20:05 +2026-08-25 19:32:16 | INFO | train rank=0 | step=42260/100000 progress=42.3% loss=0.062546 lr=6.660979e-06 elapsed=15:36:46 avg_step=1.33s recent_step=1.53s eta=21:19:54 +2026-08-25 19:32:31 | INFO | train rank=0 | step=42270/100000 progress=42.3% loss=0.074188 lr=6.659419e-06 elapsed=15:37:01 avg_step=1.33s recent_step=1.52s eta=21:19:43 +2026-08-25 19:32:42 | INFO | train rank=0 | step=42280/100000 progress=42.3% loss=0.045349 lr=6.657859e-06 elapsed=15:37:12 avg_step=1.33s recent_step=1.16s eta=21:19:28 +2026-08-25 19:32:54 | INFO | train rank=0 | step=42290/100000 progress=42.3% loss=0.050522 lr=6.656299e-06 elapsed=15:37:24 avg_step=1.33s recent_step=1.15s eta=21:19:12 +2026-08-25 19:33:10 | INFO | train rank=0 | step=42300/100000 progress=42.3% loss=0.058388 lr=6.654739e-06 elapsed=15:37:40 avg_step=1.33s recent_step=1.59s eta=21:19:02 +2026-08-25 19:33:22 | INFO | train rank=0 | step=42310/100000 progress=42.3% loss=0.068909 lr=6.653179e-06 elapsed=15:37:52 avg_step=1.33s recent_step=1.23s eta=21:18:48 +2026-08-25 19:33:37 | INFO | train rank=0 | step=42320/100000 progress=42.3% loss=0.061340 lr=6.651618e-06 elapsed=15:38:06 avg_step=1.33s recent_step=1.43s eta=21:18:36 +2026-08-25 19:33:51 | INFO | train rank=0 | step=42330/100000 progress=42.3% loss=0.047897 lr=6.650058e-06 elapsed=15:38:21 avg_step=1.33s recent_step=1.45s eta=21:18:24 +2026-08-25 19:34:05 | INFO | train rank=0 | step=42340/100000 progress=42.3% loss=0.062439 lr=6.648497e-06 elapsed=15:38:34 avg_step=1.33s recent_step=1.36s eta=21:18:11 +2026-08-25 19:34:16 | INFO | train rank=0 | step=42350/100000 progress=42.4% loss=0.046646 lr=6.646935e-06 elapsed=15:38:46 avg_step=1.33s recent_step=1.16s eta=21:17:55 +2026-08-25 19:34:30 | INFO | train rank=0 | step=42360/100000 progress=42.4% loss=0.060074 lr=6.645374e-06 elapsed=15:39:00 avg_step=1.33s recent_step=1.38s eta=21:17:43 +2026-08-25 19:34:42 | INFO | train rank=0 | step=42370/100000 progress=42.4% loss=0.042603 lr=6.643813e-06 elapsed=15:39:12 avg_step=1.33s recent_step=1.25s eta=21:17:28 +2026-08-25 19:34:54 | INFO | train rank=0 | step=42380/100000 progress=42.4% loss=0.042435 lr=6.642251e-06 elapsed=15:39:24 avg_step=1.33s recent_step=1.17s eta=21:17:13 +2026-08-25 19:35:09 | INFO | train rank=0 | step=42390/100000 progress=42.4% loss=0.044647 lr=6.640689e-06 elapsed=15:39:39 avg_step=1.33s recent_step=1.54s eta=21:17:02 +2026-08-25 19:35:21 | INFO | train rank=0 | step=42400/100000 progress=42.4% loss=0.067780 lr=6.639127e-06 elapsed=15:39:51 avg_step=1.33s recent_step=1.17s eta=21:16:47 +2026-08-25 19:35:38 | INFO | train rank=0 | step=42410/100000 progress=42.4% loss=0.061371 lr=6.637565e-06 elapsed=15:40:08 avg_step=1.33s recent_step=1.67s eta=21:16:38 +2026-08-25 19:35:52 | INFO | train rank=0 | step=42420/100000 progress=42.4% loss=0.071381 lr=6.636003e-06 elapsed=15:40:22 avg_step=1.33s recent_step=1.38s eta=21:16:26 +2026-08-25 19:36:07 | INFO | train rank=0 | step=42430/100000 progress=42.4% loss=0.057114 lr=6.634440e-06 elapsed=15:40:37 avg_step=1.33s recent_step=1.51s eta=21:16:15 +2026-08-25 19:36:18 | INFO | train rank=0 | step=42440/100000 progress=42.4% loss=0.047546 lr=6.632877e-06 elapsed=15:40:48 avg_step=1.33s recent_step=1.11s eta=21:15:59 +2026-08-25 19:36:32 | INFO | train rank=0 | step=42450/100000 progress=42.4% loss=0.058685 lr=6.631315e-06 elapsed=15:41:02 avg_step=1.33s recent_step=1.44s eta=21:15:47 +2026-08-25 19:36:44 | INFO | train rank=0 | step=42460/100000 progress=42.5% loss=0.063049 lr=6.629751e-06 elapsed=15:41:14 avg_step=1.33s recent_step=1.16s eta=21:15:31 +2026-08-25 19:36:56 | INFO | train rank=0 | step=42470/100000 progress=42.5% loss=0.047775 lr=6.628188e-06 elapsed=15:41:26 avg_step=1.33s recent_step=1.18s eta=21:15:16 +2026-08-25 19:37:10 | INFO | train rank=0 | step=42480/100000 progress=42.5% loss=0.069061 lr=6.626625e-06 elapsed=15:41:40 avg_step=1.33s recent_step=1.44s eta=21:15:04 +2026-08-25 19:37:28 | INFO | train rank=0 | step=42490/100000 progress=42.5% loss=0.035965 lr=6.625061e-06 elapsed=15:41:58 avg_step=1.33s recent_step=1.76s eta=21:14:57 +2026-08-25 19:37:39 | INFO | train rank=0 | step=42500/100000 progress=42.5% loss=0.048874 lr=6.623497e-06 elapsed=15:42:09 avg_step=1.33s recent_step=1.15s eta=21:14:41 +2026-08-25 19:37:51 | INFO | train rank=0 | step=42510/100000 progress=42.5% loss=0.062164 lr=6.621933e-06 elapsed=15:42:21 avg_step=1.33s recent_step=1.21s eta=21:14:26 +2026-08-25 19:38:03 | INFO | train rank=0 | step=42520/100000 progress=42.5% loss=0.062820 lr=6.620369e-06 elapsed=15:42:33 avg_step=1.33s recent_step=1.15s eta=21:14:10 +2026-08-25 19:38:14 | INFO | train rank=0 | step=42530/100000 progress=42.5% loss=0.074066 lr=6.618805e-06 elapsed=15:42:44 avg_step=1.33s recent_step=1.15s eta=21:13:54 +2026-08-25 19:38:27 | INFO | train rank=0 | step=42540/100000 progress=42.5% loss=0.045464 lr=6.617240e-06 elapsed=15:42:57 avg_step=1.33s recent_step=1.25s eta=21:13:40 +2026-08-25 19:38:44 | INFO | train rank=0 | step=42550/100000 progress=42.5% loss=0.053986 lr=6.615676e-06 elapsed=15:43:14 avg_step=1.33s recent_step=1.76s eta=21:13:33 +2026-08-25 19:38:56 | INFO | train rank=0 | step=42560/100000 progress=42.6% loss=0.052261 lr=6.614111e-06 elapsed=15:43:26 avg_step=1.33s recent_step=1.15s eta=21:13:17 +2026-08-25 19:39:09 | INFO | train rank=0 | step=42570/100000 progress=42.6% loss=0.051300 lr=6.612546e-06 elapsed=15:43:39 avg_step=1.33s recent_step=1.31s eta=21:13:03 +2026-08-25 19:39:28 | INFO | train rank=0 | step=42580/100000 progress=42.6% loss=0.043686 lr=6.610981e-06 elapsed=15:43:58 avg_step=1.33s recent_step=1.94s eta=21:12:58 +2026-08-25 19:39:40 | INFO | train rank=0 | step=42590/100000 progress=42.6% loss=0.065216 lr=6.609415e-06 elapsed=15:44:10 avg_step=1.33s recent_step=1.17s eta=21:12:43 +2026-08-25 19:39:53 | INFO | train rank=0 | step=42600/100000 progress=42.6% loss=0.063278 lr=6.607850e-06 elapsed=15:44:23 avg_step=1.33s recent_step=1.26s eta=21:12:28 +2026-08-25 19:40:09 | INFO | train rank=0 | step=42610/100000 progress=42.6% loss=0.045471 lr=6.606284e-06 elapsed=15:44:39 avg_step=1.33s recent_step=1.63s eta=21:12:19 +2026-08-25 19:40:25 | INFO | train rank=0 | step=42620/100000 progress=42.6% loss=0.043884 lr=6.604718e-06 elapsed=15:44:55 avg_step=1.33s recent_step=1.61s eta=21:12:09 +2026-08-25 19:40:40 | INFO | train rank=0 | step=42630/100000 progress=42.6% loss=0.037735 lr=6.603152e-06 elapsed=15:45:10 avg_step=1.33s recent_step=1.49s eta=21:11:58 +2026-08-25 19:40:51 | INFO | train rank=0 | step=42640/100000 progress=42.6% loss=0.100891 lr=6.601586e-06 elapsed=15:45:21 avg_step=1.33s recent_step=1.15s eta=21:11:43 +2026-08-25 19:41:06 | INFO | train rank=0 | step=42650/100000 progress=42.6% loss=0.069733 lr=6.600019e-06 elapsed=15:45:36 avg_step=1.33s recent_step=1.43s eta=21:11:31 +2026-08-25 19:41:18 | INFO | train rank=0 | step=42660/100000 progress=42.7% loss=0.060684 lr=6.598453e-06 elapsed=15:45:48 avg_step=1.33s recent_step=1.19s eta=21:11:15 +2026-08-25 19:41:29 | INFO | train rank=0 | step=42670/100000 progress=42.7% loss=0.043896 lr=6.596886e-06 elapsed=15:45:59 avg_step=1.33s recent_step=1.16s eta=21:11:00 +2026-08-25 19:41:46 | INFO | train rank=0 | step=42680/100000 progress=42.7% loss=0.049652 lr=6.595319e-06 elapsed=15:46:16 avg_step=1.33s recent_step=1.71s eta=21:10:52 +2026-08-25 19:42:00 | INFO | train rank=0 | step=42690/100000 progress=42.7% loss=0.044044 lr=6.593752e-06 elapsed=15:46:30 avg_step=1.33s recent_step=1.36s eta=21:10:39 +2026-08-25 19:42:15 | INFO | train rank=0 | step=42700/100000 progress=42.7% loss=0.058670 lr=6.592184e-06 elapsed=15:46:45 avg_step=1.33s recent_step=1.50s eta=21:10:28 +2026-08-25 19:42:26 | INFO | train rank=0 | step=42710/100000 progress=42.7% loss=0.131561 lr=6.590617e-06 elapsed=15:46:56 avg_step=1.33s recent_step=1.16s eta=21:10:12 +2026-08-25 19:42:38 | INFO | train rank=0 | step=42720/100000 progress=42.7% loss=0.129211 lr=6.589049e-06 elapsed=15:47:08 avg_step=1.33s recent_step=1.14s eta=21:09:56 +2026-08-25 19:42:49 | INFO | train rank=0 | step=42730/100000 progress=42.7% loss=0.037979 lr=6.587481e-06 elapsed=15:47:19 avg_step=1.33s recent_step=1.16s eta=21:09:41 +2026-08-25 19:43:02 | INFO | train rank=0 | step=42740/100000 progress=42.7% loss=0.049423 lr=6.585913e-06 elapsed=15:47:32 avg_step=1.33s recent_step=1.24s eta=21:09:26 +2026-08-25 19:43:14 | INFO | train rank=0 | step=42750/100000 progress=42.8% loss=0.061157 lr=6.584345e-06 elapsed=15:47:44 avg_step=1.33s recent_step=1.18s eta=21:09:11 +2026-08-25 19:43:25 | INFO | train rank=0 | step=42760/100000 progress=42.8% loss=0.049316 lr=6.582777e-06 elapsed=15:47:55 avg_step=1.33s recent_step=1.16s eta=21:08:55 +2026-08-25 19:43:38 | INFO | train rank=0 | step=42770/100000 progress=42.8% loss=0.060272 lr=6.581208e-06 elapsed=15:48:08 avg_step=1.33s recent_step=1.28s eta=21:08:41 +2026-08-25 19:43:53 | INFO | train rank=0 | step=42780/100000 progress=42.8% loss=0.065216 lr=6.579640e-06 elapsed=15:48:23 avg_step=1.33s recent_step=1.48s eta=21:08:30 +2026-08-25 19:44:06 | INFO | train rank=0 | step=42790/100000 progress=42.8% loss=0.048203 lr=6.578071e-06 elapsed=15:48:36 avg_step=1.33s recent_step=1.33s eta=21:08:17 +2026-08-25 19:44:20 | INFO | train rank=0 | step=42800/100000 progress=42.8% loss=0.061005 lr=6.576502e-06 elapsed=15:48:50 avg_step=1.33s recent_step=1.36s eta=21:08:04 +2026-08-25 19:44:33 | INFO | train rank=0 | step=42810/100000 progress=42.8% loss=0.059235 lr=6.574933e-06 elapsed=15:49:03 avg_step=1.33s recent_step=1.33s eta=21:07:50 +2026-08-25 19:44:49 | INFO | train rank=0 | step=42820/100000 progress=42.8% loss=0.064941 lr=6.573363e-06 elapsed=15:49:19 avg_step=1.33s recent_step=1.62s eta=21:07:41 +2026-08-25 19:45:02 | INFO | train rank=0 | step=42830/100000 progress=42.8% loss=0.061615 lr=6.571794e-06 elapsed=15:49:32 avg_step=1.33s recent_step=1.25s eta=21:07:27 +2026-08-25 19:45:16 | INFO | train rank=0 | step=42840/100000 progress=42.8% loss=0.052658 lr=6.570224e-06 elapsed=15:49:46 avg_step=1.33s recent_step=1.39s eta=21:07:14 +2026-08-25 19:45:29 | INFO | train rank=0 | step=42850/100000 progress=42.9% loss=0.053528 lr=6.568654e-06 elapsed=15:49:59 avg_step=1.33s recent_step=1.31s eta=21:07:01 +2026-08-25 19:45:41 | INFO | train rank=0 | step=42860/100000 progress=42.9% loss=0.046860 lr=6.567084e-06 elapsed=15:50:11 avg_step=1.33s recent_step=1.23s eta=21:06:46 +2026-08-25 19:45:54 | INFO | train rank=0 | step=42870/100000 progress=42.9% loss=0.062302 lr=6.565514e-06 elapsed=15:50:24 avg_step=1.33s recent_step=1.30s eta=21:06:32 +2026-08-25 19:46:09 | INFO | train rank=0 | step=42880/100000 progress=42.9% loss=0.042435 lr=6.563943e-06 elapsed=15:50:38 avg_step=1.33s recent_step=1.45s eta=21:06:21 +2026-08-25 19:46:23 | INFO | train rank=0 | step=42890/100000 progress=42.9% loss=0.050201 lr=6.562373e-06 elapsed=15:50:53 avg_step=1.33s recent_step=1.47s eta=21:06:09 +2026-08-25 19:46:35 | INFO | train rank=0 | step=42900/100000 progress=42.9% loss=0.059982 lr=6.560802e-06 elapsed=15:51:05 avg_step=1.33s recent_step=1.22s eta=21:05:54 +2026-08-25 19:46:48 | INFO | train rank=0 | step=42910/100000 progress=42.9% loss=0.066589 lr=6.559231e-06 elapsed=15:51:18 avg_step=1.33s recent_step=1.26s eta=21:05:40 +2026-08-25 19:47:02 | INFO | train rank=0 | step=42920/100000 progress=42.9% loss=0.033287 lr=6.557660e-06 elapsed=15:51:32 avg_step=1.33s recent_step=1.39s eta=21:05:28 +2026-08-25 19:47:15 | INFO | train rank=0 | step=42930/100000 progress=42.9% loss=0.070801 lr=6.556089e-06 elapsed=15:51:45 avg_step=1.33s recent_step=1.27s eta=21:05:13 +2026-08-25 19:47:27 | INFO | train rank=0 | step=42940/100000 progress=42.9% loss=0.063354 lr=6.554517e-06 elapsed=15:51:57 avg_step=1.33s recent_step=1.23s eta=21:04:59 +2026-08-25 19:47:40 | INFO | train rank=0 | step=42950/100000 progress=43.0% loss=0.059479 lr=6.552946e-06 elapsed=15:52:09 avg_step=1.33s recent_step=1.26s eta=21:04:45 +2026-08-25 19:47:53 | INFO | train rank=0 | step=42960/100000 progress=43.0% loss=0.060577 lr=6.551374e-06 elapsed=15:52:23 avg_step=1.33s recent_step=1.36s eta=21:04:32 +2026-08-25 19:48:06 | INFO | train rank=0 | step=42970/100000 progress=43.0% loss=0.062347 lr=6.549802e-06 elapsed=15:52:36 avg_step=1.33s recent_step=1.27s eta=21:04:17 +2026-08-25 19:48:19 | INFO | train rank=0 | step=42980/100000 progress=43.0% loss=0.065613 lr=6.548230e-06 elapsed=15:52:49 avg_step=1.33s recent_step=1.34s eta=21:04:04 +2026-08-25 19:48:32 | INFO | train rank=0 | step=42990/100000 progress=43.0% loss=0.040634 lr=6.546657e-06 elapsed=15:53:02 avg_step=1.33s recent_step=1.27s eta=21:03:50 +2026-08-25 19:48:46 | INFO | train rank=0 | step=43000/100000 progress=43.0% loss=0.130859 lr=6.545085e-06 elapsed=15:53:15 avg_step=1.33s recent_step=1.37s eta=21:03:37 +2026-08-25 19:49:00 | INFO | train rank=0 | step=43010/100000 progress=43.0% loss=0.035583 lr=6.543512e-06 elapsed=15:53:30 avg_step=1.33s recent_step=1.48s eta=21:03:26 +2026-08-25 19:49:13 | INFO | train rank=0 | step=43020/100000 progress=43.0% loss=0.084076 lr=6.541940e-06 elapsed=15:53:43 avg_step=1.33s recent_step=1.29s eta=21:03:12 +2026-08-25 19:49:25 | INFO | train rank=0 | step=43030/100000 progress=43.0% loss=0.039352 lr=6.540367e-06 elapsed=15:53:55 avg_step=1.33s recent_step=1.23s eta=21:02:57 +2026-08-25 19:49:40 | INFO | train rank=0 | step=43040/100000 progress=43.0% loss=0.056793 lr=6.538793e-06 elapsed=15:54:10 avg_step=1.33s recent_step=1.43s eta=21:02:46 +2026-08-25 19:49:52 | INFO | train rank=0 | step=43050/100000 progress=43.0% loss=0.056763 lr=6.537220e-06 elapsed=15:54:22 avg_step=1.33s recent_step=1.26s eta=21:02:31 +2026-08-25 19:50:06 | INFO | train rank=0 | step=43060/100000 progress=43.1% loss=0.059692 lr=6.535647e-06 elapsed=15:54:36 avg_step=1.33s recent_step=1.33s eta=21:02:18 +2026-08-25 19:50:18 | INFO | train rank=0 | step=43070/100000 progress=43.1% loss=0.085938 lr=6.534073e-06 elapsed=15:54:48 avg_step=1.33s recent_step=1.24s eta=21:02:04 +2026-08-25 19:50:37 | INFO | train rank=0 | step=43080/100000 progress=43.1% loss=0.076050 lr=6.532499e-06 elapsed=15:55:06 avg_step=1.33s recent_step=1.84s eta=21:01:57 +2026-08-25 19:50:49 | INFO | train rank=0 | step=43090/100000 progress=43.1% loss=0.092102 lr=6.530925e-06 elapsed=15:55:18 avg_step=1.33s recent_step=1.21s eta=21:01:42 +2026-08-25 19:51:01 | INFO | train rank=0 | step=43100/100000 progress=43.1% loss=0.069702 lr=6.529351e-06 elapsed=15:55:31 avg_step=1.33s recent_step=1.27s eta=21:01:28 +2026-08-25 19:51:14 | INFO | train rank=0 | step=43110/100000 progress=43.1% loss=0.049133 lr=6.527777e-06 elapsed=15:55:44 avg_step=1.33s recent_step=1.24s eta=21:01:14 +2026-08-25 19:51:27 | INFO | train rank=0 | step=43120/100000 progress=43.1% loss=0.049332 lr=6.526202e-06 elapsed=15:55:57 avg_step=1.33s recent_step=1.35s eta=21:01:00 +2026-08-25 19:51:40 | INFO | train rank=0 | step=43130/100000 progress=43.1% loss=0.046036 lr=6.524628e-06 elapsed=15:56:10 avg_step=1.33s recent_step=1.26s eta=21:00:46 +2026-08-25 19:51:52 | INFO | train rank=0 | step=43140/100000 progress=43.1% loss=0.092285 lr=6.523053e-06 elapsed=15:56:22 avg_step=1.33s recent_step=1.24s eta=21:00:32 +2026-08-25 19:52:05 | INFO | train rank=0 | step=43150/100000 progress=43.1% loss=0.042862 lr=6.521478e-06 elapsed=15:56:35 avg_step=1.33s recent_step=1.30s eta=21:00:18 +2026-08-25 19:52:19 | INFO | train rank=0 | step=43160/100000 progress=43.2% loss=0.051208 lr=6.519903e-06 elapsed=15:56:49 avg_step=1.33s recent_step=1.35s eta=21:00:05 +2026-08-25 19:52:32 | INFO | train rank=0 | step=43170/100000 progress=43.2% loss=0.047043 lr=6.518327e-06 elapsed=15:57:02 avg_step=1.33s recent_step=1.29s eta=20:59:51 +2026-08-25 19:52:44 | INFO | train rank=0 | step=43180/100000 progress=43.2% loss=0.038666 lr=6.516752e-06 elapsed=15:57:14 avg_step=1.33s recent_step=1.24s eta=20:59:37 +2026-08-25 19:52:59 | INFO | train rank=0 | step=43190/100000 progress=43.2% loss=0.071838 lr=6.515176e-06 elapsed=15:57:29 avg_step=1.33s recent_step=1.45s eta=20:59:25 +2026-08-25 19:53:12 | INFO | train rank=0 | step=43200/100000 progress=43.2% loss=0.127808 lr=6.513601e-06 elapsed=15:57:42 avg_step=1.33s recent_step=1.32s eta=20:59:12 +2026-08-25 19:53:25 | INFO | train rank=0 | step=43210/100000 progress=43.2% loss=0.074493 lr=6.512025e-06 elapsed=15:57:55 avg_step=1.33s recent_step=1.31s eta=20:58:58 +2026-08-25 19:53:37 | INFO | train rank=0 | step=43220/100000 progress=43.2% loss=0.080688 lr=6.510448e-06 elapsed=15:58:07 avg_step=1.33s recent_step=1.24s eta=20:58:44 +2026-08-25 19:53:50 | INFO | train rank=0 | step=43230/100000 progress=43.2% loss=0.041054 lr=6.508872e-06 elapsed=15:58:20 avg_step=1.33s recent_step=1.26s eta=20:58:29 +2026-08-25 19:54:03 | INFO | train rank=0 | step=43240/100000 progress=43.2% loss=0.071533 lr=6.507296e-06 elapsed=15:58:33 avg_step=1.33s recent_step=1.34s eta=20:58:16 +2026-08-25 19:54:17 | INFO | train rank=0 | step=43250/100000 progress=43.2% loss=0.044739 lr=6.505719e-06 elapsed=15:58:47 avg_step=1.33s recent_step=1.36s eta=20:58:03 +2026-08-25 19:54:28 | INFO | train rank=0 | step=43260/100000 progress=43.3% loss=0.101746 lr=6.504142e-06 elapsed=15:58:58 avg_step=1.33s recent_step=1.14s eta=20:57:47 +2026-08-25 19:54:41 | INFO | train rank=0 | step=43270/100000 progress=43.3% loss=0.045502 lr=6.502565e-06 elapsed=15:59:11 avg_step=1.33s recent_step=1.25s eta=20:57:33 +2026-08-25 19:54:57 | INFO | train rank=0 | step=43280/100000 progress=43.3% loss=0.067505 lr=6.500988e-06 elapsed=15:59:27 avg_step=1.33s recent_step=1.61s eta=20:57:24 +2026-08-25 19:55:09 | INFO | train rank=0 | step=43290/100000 progress=43.3% loss=0.061295 lr=6.499411e-06 elapsed=15:59:38 avg_step=1.33s recent_step=1.16s eta=20:57:08 +2026-08-25 19:55:22 | INFO | train rank=0 | step=43300/100000 progress=43.3% loss=0.087097 lr=6.497833e-06 elapsed=15:59:52 avg_step=1.33s recent_step=1.34s eta=20:56:55 +2026-08-25 19:55:34 | INFO | train rank=0 | step=43310/100000 progress=43.3% loss=0.057190 lr=6.496256e-06 elapsed=16:00:04 avg_step=1.33s recent_step=1.18s eta=20:56:40 +2026-08-25 19:55:50 | INFO | train rank=0 | step=43320/100000 progress=43.3% loss=0.039108 lr=6.494678e-06 elapsed=16:00:20 avg_step=1.33s recent_step=1.63s eta=20:56:30 +2026-08-25 19:56:02 | INFO | train rank=0 | step=43330/100000 progress=43.3% loss=0.059143 lr=6.493100e-06 elapsed=16:00:31 avg_step=1.33s recent_step=1.15s eta=20:56:15 +2026-08-25 19:56:14 | INFO | train rank=0 | step=43340/100000 progress=43.3% loss=0.048126 lr=6.491522e-06 elapsed=16:00:44 avg_step=1.33s recent_step=1.24s eta=20:56:00 +2026-08-25 19:56:30 | INFO | train rank=0 | step=43350/100000 progress=43.4% loss=0.064056 lr=6.489944e-06 elapsed=16:01:00 avg_step=1.33s recent_step=1.60s eta=20:55:50 +2026-08-25 19:56:44 | INFO | train rank=0 | step=43360/100000 progress=43.4% loss=0.072113 lr=6.488365e-06 elapsed=16:01:14 avg_step=1.33s recent_step=1.41s eta=20:55:38 +2026-08-25 19:56:57 | INFO | train rank=0 | step=43370/100000 progress=43.4% loss=0.079712 lr=6.486787e-06 elapsed=16:01:26 avg_step=1.33s recent_step=1.25s eta=20:55:24 +2026-08-25 19:57:12 | INFO | train rank=0 | step=43380/100000 progress=43.4% loss=0.053131 lr=6.485208e-06 elapsed=16:01:42 avg_step=1.33s recent_step=1.57s eta=20:55:14 +2026-08-25 19:57:25 | INFO | train rank=0 | step=43390/100000 progress=43.4% loss=0.050171 lr=6.483629e-06 elapsed=16:01:55 avg_step=1.33s recent_step=1.27s eta=20:54:59 +2026-08-25 19:57:38 | INFO | train rank=0 | step=43400/100000 progress=43.4% loss=0.063522 lr=6.482050e-06 elapsed=16:02:08 avg_step=1.33s recent_step=1.34s eta=20:54:46 +2026-08-25 19:57:51 | INFO | train rank=0 | step=43410/100000 progress=43.4% loss=0.048584 lr=6.480471e-06 elapsed=16:02:21 avg_step=1.33s recent_step=1.30s eta=20:54:32 +2026-08-25 19:58:06 | INFO | train rank=0 | step=43420/100000 progress=43.4% loss=0.035126 lr=6.478891e-06 elapsed=16:02:36 avg_step=1.33s recent_step=1.49s eta=20:54:21 +2026-08-25 19:58:19 | INFO | train rank=0 | step=43430/100000 progress=43.4% loss=0.046295 lr=6.477312e-06 elapsed=16:02:49 avg_step=1.33s recent_step=1.26s eta=20:54:07 +2026-08-25 19:58:32 | INFO | train rank=0 | step=43440/100000 progress=43.4% loss=0.074677 lr=6.475732e-06 elapsed=16:03:02 avg_step=1.33s recent_step=1.35s eta=20:53:54 +2026-08-25 19:58:51 | INFO | train rank=0 | step=43450/100000 progress=43.5% loss=0.065857 lr=6.474152e-06 elapsed=16:03:21 avg_step=1.33s recent_step=1.87s eta=20:53:48 +2026-08-25 19:59:04 | INFO | train rank=0 | step=43460/100000 progress=43.5% loss=0.047638 lr=6.472572e-06 elapsed=16:03:33 avg_step=1.33s recent_step=1.26s eta=20:53:33 +2026-08-25 19:59:16 | INFO | train rank=0 | step=43470/100000 progress=43.5% loss=0.069946 lr=6.470992e-06 elapsed=16:03:46 avg_step=1.33s recent_step=1.26s eta=20:53:19 +2026-08-25 19:59:30 | INFO | train rank=0 | step=43480/100000 progress=43.5% loss=0.050407 lr=6.469411e-06 elapsed=16:04:00 avg_step=1.33s recent_step=1.36s eta=20:53:06 +2026-08-25 19:59:42 | INFO | train rank=0 | step=43490/100000 progress=43.5% loss=0.145569 lr=6.467831e-06 elapsed=16:04:12 avg_step=1.33s recent_step=1.23s eta=20:52:52 +2026-08-25 19:59:54 | INFO | train rank=0 | step=43500/100000 progress=43.5% loss=0.038986 lr=6.466250e-06 elapsed=16:04:24 avg_step=1.33s recent_step=1.26s eta=20:52:37 +2026-08-25 20:00:07 | INFO | train rank=0 | step=43510/100000 progress=43.5% loss=0.067596 lr=6.464669e-06 elapsed=16:04:37 avg_step=1.33s recent_step=1.27s eta=20:52:23 +2026-08-25 20:00:21 | INFO | train rank=0 | step=43520/100000 progress=43.5% loss=0.031975 lr=6.463088e-06 elapsed=16:04:51 avg_step=1.33s recent_step=1.36s eta=20:52:10 +2026-08-25 20:00:33 | INFO | train rank=0 | step=43530/100000 progress=43.5% loss=0.056396 lr=6.461507e-06 elapsed=16:05:03 avg_step=1.33s recent_step=1.20s eta=20:51:55 +2026-08-25 20:00:45 | INFO | train rank=0 | step=43540/100000 progress=43.5% loss=0.044861 lr=6.459926e-06 elapsed=16:05:15 avg_step=1.33s recent_step=1.20s eta=20:51:40 +2026-08-25 20:00:58 | INFO | train rank=0 | step=43550/100000 progress=43.5% loss=0.039001 lr=6.458344e-06 elapsed=16:05:27 avg_step=1.33s recent_step=1.29s eta=20:51:26 +2026-08-25 20:01:10 | INFO | train rank=0 | step=43560/100000 progress=43.6% loss=0.068298 lr=6.456763e-06 elapsed=16:05:40 avg_step=1.33s recent_step=1.24s eta=20:51:12 +2026-08-25 20:01:23 | INFO | train rank=0 | step=43570/100000 progress=43.6% loss=0.045319 lr=6.455181e-06 elapsed=16:05:53 avg_step=1.33s recent_step=1.31s eta=20:50:58 +2026-08-25 20:01:40 | INFO | train rank=0 | step=43580/100000 progress=43.6% loss=0.055878 lr=6.453599e-06 elapsed=16:06:10 avg_step=1.33s recent_step=1.67s eta=20:50:49 +2026-08-25 20:01:52 | INFO | train rank=0 | step=43590/100000 progress=43.6% loss=0.062500 lr=6.452017e-06 elapsed=16:06:22 avg_step=1.33s recent_step=1.25s eta=20:50:35 +2026-08-25 20:02:05 | INFO | train rank=0 | step=43600/100000 progress=43.6% loss=0.041214 lr=6.450434e-06 elapsed=16:06:34 avg_step=1.33s recent_step=1.22s eta=20:50:20 +2026-08-25 20:02:17 | INFO | train rank=0 | step=43610/100000 progress=43.6% loss=0.048660 lr=6.448852e-06 elapsed=16:06:47 avg_step=1.33s recent_step=1.26s eta=20:50:06 +2026-08-25 20:02:30 | INFO | train rank=0 | step=43620/100000 progress=43.6% loss=0.049164 lr=6.447269e-06 elapsed=16:07:00 avg_step=1.33s recent_step=1.30s eta=20:49:53 +2026-08-25 20:02:43 | INFO | train rank=0 | step=43630/100000 progress=43.6% loss=0.066254 lr=6.445687e-06 elapsed=16:07:12 avg_step=1.33s recent_step=1.24s eta=20:49:38 +2026-08-25 20:02:56 | INFO | train rank=0 | step=43640/100000 progress=43.6% loss=0.058121 lr=6.444104e-06 elapsed=16:07:26 avg_step=1.33s recent_step=1.38s eta=20:49:25 +2026-08-25 20:03:08 | INFO | train rank=0 | step=43650/100000 progress=43.6% loss=0.052017 lr=6.442521e-06 elapsed=16:07:38 avg_step=1.33s recent_step=1.16s eta=20:49:10 +2026-08-25 20:03:20 | INFO | train rank=0 | step=43660/100000 progress=43.7% loss=0.060577 lr=6.440937e-06 elapsed=16:07:50 avg_step=1.33s recent_step=1.20s eta=20:48:55 +2026-08-25 20:03:36 | INFO | train rank=0 | step=43670/100000 progress=43.7% loss=0.064087 lr=6.439354e-06 elapsed=16:08:06 avg_step=1.33s recent_step=1.62s eta=20:48:45 +2026-08-25 20:03:53 | INFO | train rank=0 | step=43680/100000 progress=43.7% loss=0.070221 lr=6.437770e-06 elapsed=16:08:23 avg_step=1.33s recent_step=1.73s eta=20:48:37 +2026-08-25 20:04:06 | INFO | train rank=0 | step=43690/100000 progress=43.7% loss=0.050568 lr=6.436187e-06 elapsed=16:08:35 avg_step=1.33s recent_step=1.21s eta=20:48:22 +2026-08-25 20:04:21 | INFO | train rank=0 | step=43700/100000 progress=43.7% loss=0.069061 lr=6.434603e-06 elapsed=16:08:51 avg_step=1.33s recent_step=1.57s eta=20:48:12 +2026-08-25 20:04:35 | INFO | train rank=0 | step=43710/100000 progress=43.7% loss=0.041992 lr=6.433019e-06 elapsed=16:09:05 avg_step=1.33s recent_step=1.39s eta=20:48:00 +2026-08-25 20:04:47 | INFO | train rank=0 | step=43720/100000 progress=43.7% loss=0.059937 lr=6.431435e-06 elapsed=16:09:17 avg_step=1.33s recent_step=1.20s eta=20:47:45 +2026-08-25 20:05:02 | INFO | train rank=0 | step=43730/100000 progress=43.7% loss=0.117920 lr=6.429850e-06 elapsed=16:09:32 avg_step=1.33s recent_step=1.50s eta=20:47:33 +2026-08-25 20:05:15 | INFO | train rank=0 | step=43740/100000 progress=43.7% loss=0.065338 lr=6.428266e-06 elapsed=16:09:45 avg_step=1.33s recent_step=1.28s eta=20:47:20 +2026-08-25 20:05:27 | INFO | train rank=0 | step=43750/100000 progress=43.8% loss=0.033905 lr=6.426681e-06 elapsed=16:09:57 avg_step=1.33s recent_step=1.19s eta=20:47:04 +2026-08-25 20:05:41 | INFO | train rank=0 | step=43760/100000 progress=43.8% loss=0.039917 lr=6.425096e-06 elapsed=16:10:11 avg_step=1.33s recent_step=1.39s eta=20:46:52 +2026-08-25 20:05:55 | INFO | train rank=0 | step=43770/100000 progress=43.8% loss=0.061249 lr=6.423511e-06 elapsed=16:10:25 avg_step=1.33s recent_step=1.41s eta=20:46:40 +2026-08-25 20:06:14 | INFO | train rank=0 | step=43780/100000 progress=43.8% loss=0.109436 lr=6.421926e-06 elapsed=16:10:44 avg_step=1.33s recent_step=1.97s eta=20:46:34 +2026-08-25 20:06:26 | INFO | train rank=0 | step=43790/100000 progress=43.8% loss=0.032928 lr=6.420341e-06 elapsed=16:10:56 avg_step=1.33s recent_step=1.15s eta=20:46:19 +2026-08-25 20:06:38 | INFO | train rank=0 | step=43800/100000 progress=43.8% loss=0.056473 lr=6.418756e-06 elapsed=16:11:08 avg_step=1.33s recent_step=1.17s eta=20:46:04 +2026-08-25 20:06:50 | INFO | train rank=0 | step=43810/100000 progress=43.8% loss=0.036438 lr=6.417170e-06 elapsed=16:11:20 avg_step=1.33s recent_step=1.21s eta=20:45:49 +2026-08-25 20:07:03 | INFO | train rank=0 | step=43820/100000 progress=43.8% loss=0.108398 lr=6.415584e-06 elapsed=16:11:33 avg_step=1.33s recent_step=1.36s eta=20:45:36 +2026-08-25 20:07:19 | INFO | train rank=0 | step=43830/100000 progress=43.8% loss=0.057739 lr=6.413998e-06 elapsed=16:11:49 avg_step=1.33s recent_step=1.57s eta=20:45:26 +2026-08-25 20:07:34 | INFO | train rank=0 | step=43840/100000 progress=43.8% loss=0.052155 lr=6.412412e-06 elapsed=16:12:04 avg_step=1.33s recent_step=1.50s eta=20:45:15 +2026-08-25 20:07:46 | INFO | train rank=0 | step=43850/100000 progress=43.9% loss=0.052078 lr=6.410826e-06 elapsed=16:12:16 avg_step=1.33s recent_step=1.20s eta=20:45:00 +2026-08-25 20:08:04 | INFO | train rank=0 | step=43860/100000 progress=43.9% loss=0.034538 lr=6.409240e-06 elapsed=16:12:34 avg_step=1.33s recent_step=1.80s eta=20:44:52 +2026-08-25 20:08:18 | INFO | train rank=0 | step=43870/100000 progress=43.9% loss=0.063339 lr=6.407653e-06 elapsed=16:12:48 avg_step=1.33s recent_step=1.41s eta=20:44:40 +2026-08-25 20:08:31 | INFO | train rank=0 | step=43880/100000 progress=43.9% loss=0.053360 lr=6.406066e-06 elapsed=16:13:01 avg_step=1.33s recent_step=1.32s eta=20:44:26 +2026-08-25 20:08:44 | INFO | train rank=0 | step=43890/100000 progress=43.9% loss=0.080719 lr=6.404480e-06 elapsed=16:13:13 avg_step=1.33s recent_step=1.22s eta=20:44:12 +2026-08-25 20:08:55 | INFO | train rank=0 | step=43900/100000 progress=43.9% loss=0.056671 lr=6.402893e-06 elapsed=16:13:25 avg_step=1.33s recent_step=1.17s eta=20:43:56 +2026-08-25 20:09:07 | INFO | train rank=0 | step=43910/100000 progress=43.9% loss=0.069519 lr=6.401306e-06 elapsed=16:13:37 avg_step=1.33s recent_step=1.21s eta=20:43:42 +2026-08-25 20:09:20 | INFO | train rank=0 | step=43920/100000 progress=43.9% loss=0.073639 lr=6.399718e-06 elapsed=16:13:50 avg_step=1.33s recent_step=1.28s eta=20:43:28 +2026-08-25 20:09:32 | INFO | train rank=0 | step=43930/100000 progress=43.9% loss=0.068817 lr=6.398131e-06 elapsed=16:14:02 avg_step=1.33s recent_step=1.21s eta=20:43:13 +2026-08-25 20:09:47 | INFO | train rank=0 | step=43940/100000 progress=43.9% loss=0.055511 lr=6.396543e-06 elapsed=16:14:17 avg_step=1.33s recent_step=1.45s eta=20:43:01 +2026-08-25 20:09:59 | INFO | train rank=0 | step=43950/100000 progress=44.0% loss=0.062393 lr=6.394956e-06 elapsed=16:14:29 avg_step=1.33s recent_step=1.21s eta=20:42:46 +2026-08-25 20:10:12 | INFO | train rank=0 | step=43960/100000 progress=44.0% loss=0.021248 lr=6.393368e-06 elapsed=16:14:42 avg_step=1.33s recent_step=1.30s eta=20:42:32 +2026-08-25 20:10:24 | INFO | train rank=0 | step=43970/100000 progress=44.0% loss=0.056427 lr=6.391780e-06 elapsed=16:14:54 avg_step=1.33s recent_step=1.23s eta=20:42:18 +2026-08-25 20:10:42 | INFO | train rank=0 | step=43980/100000 progress=44.0% loss=0.054871 lr=6.390191e-06 elapsed=16:15:12 avg_step=1.33s recent_step=1.80s eta=20:42:11 +2026-08-25 20:10:54 | INFO | train rank=0 | step=43990/100000 progress=44.0% loss=0.071960 lr=6.388603e-06 elapsed=16:15:24 avg_step=1.33s recent_step=1.15s eta=20:41:55 +2026-08-25 20:11:06 | INFO | train rank=0 | step=44000/100000 progress=44.0% loss=0.068253 lr=6.387015e-06 elapsed=16:15:35 avg_step=1.33s recent_step=1.17s eta=20:41:40 +2026-08-25 20:11:18 | INFO | train rank=0 | step=44010/100000 progress=44.0% loss=0.034698 lr=6.385426e-06 elapsed=16:15:47 avg_step=1.33s recent_step=1.20s eta=20:41:25 +2026-08-25 20:11:31 | INFO | train rank=0 | step=44020/100000 progress=44.0% loss=0.057495 lr=6.383837e-06 elapsed=16:16:01 avg_step=1.33s recent_step=1.38s eta=20:41:12 +2026-08-25 20:11:43 | INFO | train rank=0 | step=44030/100000 progress=44.0% loss=0.059906 lr=6.382248e-06 elapsed=16:16:13 avg_step=1.33s recent_step=1.18s eta=20:40:57 +2026-08-25 20:11:56 | INFO | train rank=0 | step=44040/100000 progress=44.0% loss=0.058533 lr=6.380659e-06 elapsed=16:16:26 avg_step=1.33s recent_step=1.27s eta=20:40:43 +2026-08-25 20:12:14 | INFO | train rank=0 | step=44050/100000 progress=44.0% loss=0.061127 lr=6.379070e-06 elapsed=16:16:44 avg_step=1.33s recent_step=1.80s eta=20:40:35 +2026-08-25 20:12:28 | INFO | train rank=0 | step=44060/100000 progress=44.1% loss=0.083008 lr=6.377480e-06 elapsed=16:16:58 avg_step=1.33s recent_step=1.46s eta=20:40:24 +2026-08-25 20:12:40 | INFO | train rank=0 | step=44070/100000 progress=44.1% loss=0.036850 lr=6.375891e-06 elapsed=16:17:10 avg_step=1.33s recent_step=1.19s eta=20:40:09 +2026-08-25 20:12:52 | INFO | train rank=0 | step=44080/100000 progress=44.1% loss=0.072205 lr=6.374301e-06 elapsed=16:17:21 avg_step=1.33s recent_step=1.13s eta=20:39:53 +2026-08-25 20:13:04 | INFO | train rank=0 | step=44090/100000 progress=44.1% loss=0.030693 lr=6.372711e-06 elapsed=16:17:34 avg_step=1.33s recent_step=1.21s eta=20:39:38 +2026-08-25 20:13:16 | INFO | train rank=0 | step=44100/100000 progress=44.1% loss=0.051483 lr=6.371121e-06 elapsed=16:17:46 avg_step=1.33s recent_step=1.26s eta=20:39:24 +2026-08-25 20:13:30 | INFO | train rank=0 | step=44110/100000 progress=44.1% loss=0.069672 lr=6.369531e-06 elapsed=16:18:00 avg_step=1.33s recent_step=1.36s eta=20:39:11 +2026-08-25 20:13:42 | INFO | train rank=0 | step=44120/100000 progress=44.1% loss=0.043930 lr=6.367941e-06 elapsed=16:18:12 avg_step=1.33s recent_step=1.19s eta=20:38:56 +2026-08-25 20:13:59 | INFO | train rank=0 | step=44130/100000 progress=44.1% loss=0.042709 lr=6.366350e-06 elapsed=16:18:29 avg_step=1.33s recent_step=1.76s eta=20:38:48 +2026-08-25 20:14:11 | INFO | train rank=0 | step=44140/100000 progress=44.1% loss=0.062744 lr=6.364760e-06 elapsed=16:18:41 avg_step=1.33s recent_step=1.14s eta=20:38:32 +2026-08-25 20:14:23 | INFO | train rank=0 | step=44150/100000 progress=44.1% loss=0.032242 lr=6.363169e-06 elapsed=16:18:53 avg_step=1.33s recent_step=1.20s eta=20:38:17 +2026-08-25 20:14:38 | INFO | train rank=0 | step=44160/100000 progress=44.2% loss=0.045540 lr=6.361578e-06 elapsed=16:19:08 avg_step=1.33s recent_step=1.54s eta=20:38:07 +2026-08-25 20:14:54 | INFO | train rank=0 | step=44170/100000 progress=44.2% loss=0.024384 lr=6.359987e-06 elapsed=16:19:24 avg_step=1.33s recent_step=1.57s eta=20:37:56 +2026-08-25 20:15:07 | INFO | train rank=0 | step=44180/100000 progress=44.2% loss=0.038437 lr=6.358396e-06 elapsed=16:19:37 avg_step=1.33s recent_step=1.31s eta=20:37:43 +2026-08-25 20:15:18 | INFO | train rank=0 | step=44190/100000 progress=44.2% loss=0.041580 lr=6.356804e-06 elapsed=16:19:48 avg_step=1.33s recent_step=1.15s eta=20:37:27 +2026-08-25 20:15:32 | INFO | train rank=0 | step=44200/100000 progress=44.2% loss=0.038040 lr=6.355213e-06 elapsed=16:20:02 avg_step=1.33s recent_step=1.39s eta=20:37:15 +2026-08-25 20:15:48 | INFO | train rank=0 | step=44210/100000 progress=44.2% loss=0.065033 lr=6.353621e-06 elapsed=16:20:18 avg_step=1.33s recent_step=1.61s eta=20:37:05 +2026-08-25 20:16:00 | INFO | train rank=0 | step=44220/100000 progress=44.2% loss=0.047165 lr=6.352029e-06 elapsed=16:20:30 avg_step=1.33s recent_step=1.20s eta=20:36:50 +2026-08-25 20:16:12 | INFO | train rank=0 | step=44230/100000 progress=44.2% loss=0.049423 lr=6.350438e-06 elapsed=16:20:42 avg_step=1.33s recent_step=1.21s eta=20:36:35 +2026-08-25 20:16:28 | INFO | train rank=0 | step=44240/100000 progress=44.2% loss=0.054153 lr=6.348845e-06 elapsed=16:20:58 avg_step=1.33s recent_step=1.56s eta=20:36:25 +2026-08-25 20:16:40 | INFO | train rank=0 | step=44250/100000 progress=44.2% loss=0.031059 lr=6.347253e-06 elapsed=16:21:09 avg_step=1.33s recent_step=1.15s eta=20:36:09 +2026-08-25 20:16:56 | INFO | train rank=0 | step=44260/100000 progress=44.3% loss=0.038109 lr=6.345661e-06 elapsed=16:21:26 avg_step=1.33s recent_step=1.67s eta=20:36:00 +2026-08-25 20:17:08 | INFO | train rank=0 | step=44270/100000 progress=44.3% loss=0.038956 lr=6.344068e-06 elapsed=16:21:38 avg_step=1.33s recent_step=1.15s eta=20:35:44 +2026-08-25 20:17:21 | INFO | train rank=0 | step=44280/100000 progress=44.3% loss=0.053711 lr=6.342476e-06 elapsed=16:21:51 avg_step=1.33s recent_step=1.31s eta=20:35:31 +2026-08-25 20:17:34 | INFO | train rank=0 | step=44290/100000 progress=44.3% loss=0.054382 lr=6.340883e-06 elapsed=16:22:03 avg_step=1.33s recent_step=1.27s eta=20:35:17 +2026-08-25 20:17:51 | INFO | train rank=0 | step=44300/100000 progress=44.3% loss=0.058273 lr=6.339290e-06 elapsed=16:22:21 avg_step=1.33s recent_step=1.76s eta=20:35:09 +2026-08-25 20:18:03 | INFO | train rank=0 | step=44310/100000 progress=44.3% loss=0.092804 lr=6.337697e-06 elapsed=16:22:33 avg_step=1.33s recent_step=1.15s eta=20:34:53 +2026-08-25 20:18:19 | INFO | train rank=0 | step=44320/100000 progress=44.3% loss=0.049042 lr=6.336103e-06 elapsed=16:22:49 avg_step=1.33s recent_step=1.62s eta=20:34:44 +2026-08-25 20:18:31 | INFO | train rank=0 | step=44330/100000 progress=44.3% loss=0.042038 lr=6.334510e-06 elapsed=16:23:01 avg_step=1.33s recent_step=1.18s eta=20:34:28 +2026-08-25 20:18:43 | INFO | train rank=0 | step=44340/100000 progress=44.3% loss=0.060867 lr=6.332916e-06 elapsed=16:23:13 avg_step=1.33s recent_step=1.20s eta=20:34:13 +2026-08-25 20:18:54 | INFO | train rank=0 | step=44350/100000 progress=44.4% loss=0.043777 lr=6.331323e-06 elapsed=16:23:24 avg_step=1.33s recent_step=1.14s eta=20:33:58 +2026-08-25 20:19:06 | INFO | train rank=0 | step=44360/100000 progress=44.4% loss=0.097534 lr=6.329729e-06 elapsed=16:23:36 avg_step=1.33s recent_step=1.17s eta=20:33:42 +2026-08-25 20:19:17 | INFO | train rank=0 | step=44370/100000 progress=44.4% loss=0.056656 lr=6.328135e-06 elapsed=16:23:47 avg_step=1.33s recent_step=1.18s eta=20:33:27 +2026-08-25 20:19:31 | INFO | train rank=0 | step=44380/100000 progress=44.4% loss=0.031097 lr=6.326541e-06 elapsed=16:24:01 avg_step=1.33s recent_step=1.37s eta=20:33:14 +2026-08-25 20:19:45 | INFO | train rank=0 | step=44390/100000 progress=44.4% loss=0.055695 lr=6.324946e-06 elapsed=16:24:15 avg_step=1.33s recent_step=1.37s eta=20:33:02 +2026-08-25 20:19:57 | INFO | train rank=0 | step=44400/100000 progress=44.4% loss=0.060944 lr=6.323352e-06 elapsed=16:24:27 avg_step=1.33s recent_step=1.17s eta=20:32:46 +2026-08-25 20:20:11 | INFO | train rank=0 | step=44410/100000 progress=44.4% loss=0.047012 lr=6.321757e-06 elapsed=16:24:41 avg_step=1.33s recent_step=1.45s eta=20:32:35 +2026-08-25 20:20:25 | INFO | train rank=0 | step=44420/100000 progress=44.4% loss=0.059906 lr=6.320163e-06 elapsed=16:24:55 avg_step=1.33s recent_step=1.39s eta=20:32:22 +2026-08-25 20:20:37 | INFO | train rank=0 | step=44430/100000 progress=44.4% loss=0.039742 lr=6.318568e-06 elapsed=16:25:06 avg_step=1.33s recent_step=1.15s eta=20:32:06 +2026-08-25 20:20:49 | INFO | train rank=0 | step=44440/100000 progress=44.4% loss=0.062195 lr=6.316973e-06 elapsed=16:25:18 avg_step=1.33s recent_step=1.21s eta=20:31:52 +2026-08-25 20:21:00 | INFO | train rank=0 | step=44450/100000 progress=44.5% loss=0.042542 lr=6.315378e-06 elapsed=16:25:30 avg_step=1.33s recent_step=1.17s eta=20:31:36 +2026-08-25 20:21:12 | INFO | train rank=0 | step=44460/100000 progress=44.5% loss=0.035782 lr=6.313782e-06 elapsed=16:25:42 avg_step=1.33s recent_step=1.14s eta=20:31:21 +2026-08-25 20:21:27 | INFO | train rank=0 | step=44470/100000 progress=44.5% loss=0.054672 lr=6.312187e-06 elapsed=16:25:57 avg_step=1.33s recent_step=1.50s eta=20:31:09 +2026-08-25 20:21:40 | INFO | train rank=0 | step=44480/100000 progress=44.5% loss=0.062103 lr=6.310591e-06 elapsed=16:26:10 avg_step=1.33s recent_step=1.37s eta=20:30:57 +2026-08-25 20:21:52 | INFO | train rank=0 | step=44490/100000 progress=44.5% loss=0.077850 lr=6.308996e-06 elapsed=16:26:22 avg_step=1.33s recent_step=1.15s eta=20:30:41 +2026-08-25 20:22:08 | INFO | train rank=0 | step=44500/100000 progress=44.5% loss=0.052460 lr=6.307400e-06 elapsed=16:26:38 avg_step=1.33s recent_step=1.62s eta=20:30:31 +2026-08-25 20:22:24 | INFO | train rank=0 | step=44510/100000 progress=44.5% loss=0.058105 lr=6.305804e-06 elapsed=16:26:54 avg_step=1.33s recent_step=1.62s eta=20:30:22 +2026-08-25 20:22:40 | INFO | train rank=0 | step=44520/100000 progress=44.5% loss=0.081848 lr=6.304208e-06 elapsed=16:27:10 avg_step=1.33s recent_step=1.60s eta=20:30:12 +2026-08-25 20:22:52 | INFO | train rank=0 | step=44530/100000 progress=44.5% loss=0.066376 lr=6.302611e-06 elapsed=16:27:22 avg_step=1.33s recent_step=1.15s eta=20:29:56 +2026-08-25 20:23:04 | INFO | train rank=0 | step=44540/100000 progress=44.5% loss=0.043396 lr=6.301015e-06 elapsed=16:27:34 avg_step=1.33s recent_step=1.24s eta=20:29:42 +2026-08-25 20:23:16 | INFO | train rank=0 | step=44550/100000 progress=44.5% loss=0.047318 lr=6.299418e-06 elapsed=16:27:46 avg_step=1.33s recent_step=1.21s eta=20:29:27 +2026-08-25 20:23:28 | INFO | train rank=0 | step=44560/100000 progress=44.6% loss=0.070114 lr=6.297821e-06 elapsed=16:27:58 avg_step=1.33s recent_step=1.17s eta=20:29:12 +2026-08-25 20:23:40 | INFO | train rank=0 | step=44570/100000 progress=44.6% loss=0.039124 lr=6.296225e-06 elapsed=16:28:10 avg_step=1.33s recent_step=1.20s eta=20:28:57 +2026-08-25 20:24:00 | INFO | train rank=0 | step=44580/100000 progress=44.6% loss=0.045090 lr=6.294628e-06 elapsed=16:28:30 avg_step=1.33s recent_step=2.01s eta=20:28:52 +2026-08-25 20:24:11 | INFO | train rank=0 | step=44590/100000 progress=44.6% loss=0.047195 lr=6.293030e-06 elapsed=16:28:41 avg_step=1.33s recent_step=1.13s eta=20:28:36 +2026-08-25 20:24:23 | INFO | train rank=0 | step=44600/100000 progress=44.6% loss=0.058472 lr=6.291433e-06 elapsed=16:28:53 avg_step=1.33s recent_step=1.22s eta=20:28:21 +2026-08-25 20:24:39 | INFO | train rank=0 | step=44610/100000 progress=44.6% loss=0.053017 lr=6.289836e-06 elapsed=16:29:09 avg_step=1.33s recent_step=1.56s eta=20:28:11 +2026-08-25 20:24:50 | INFO | train rank=0 | step=44620/100000 progress=44.6% loss=0.045502 lr=6.288238e-06 elapsed=16:29:20 avg_step=1.33s recent_step=1.13s eta=20:27:55 +2026-08-25 20:25:05 | INFO | train rank=0 | step=44630/100000 progress=44.6% loss=0.056137 lr=6.286640e-06 elapsed=16:29:35 avg_step=1.33s recent_step=1.47s eta=20:27:43 +2026-08-25 20:25:17 | INFO | train rank=0 | step=44640/100000 progress=44.6% loss=0.058014 lr=6.285043e-06 elapsed=16:29:46 avg_step=1.33s recent_step=1.14s eta=20:27:28 +2026-08-25 20:25:30 | INFO | train rank=0 | step=44650/100000 progress=44.6% loss=0.046814 lr=6.283445e-06 elapsed=16:30:00 avg_step=1.33s recent_step=1.36s eta=20:27:15 +2026-08-25 20:25:45 | INFO | train rank=0 | step=44660/100000 progress=44.7% loss=0.088684 lr=6.281846e-06 elapsed=16:30:15 avg_step=1.33s recent_step=1.53s eta=20:27:04 +2026-08-25 20:25:57 | INFO | train rank=0 | step=44670/100000 progress=44.7% loss=0.055756 lr=6.280248e-06 elapsed=16:30:26 avg_step=1.33s recent_step=1.13s eta=20:26:48 +2026-08-25 20:26:09 | INFO | train rank=0 | step=44680/100000 progress=44.7% loss=0.043091 lr=6.278650e-06 elapsed=16:30:39 avg_step=1.33s recent_step=1.29s eta=20:26:34 +2026-08-25 20:26:25 | INFO | train rank=0 | step=44690/100000 progress=44.7% loss=0.035294 lr=6.277051e-06 elapsed=16:30:55 avg_step=1.33s recent_step=1.60s eta=20:26:24 +2026-08-25 20:26:37 | INFO | train rank=0 | step=44700/100000 progress=44.7% loss=0.060059 lr=6.275452e-06 elapsed=16:31:07 avg_step=1.33s recent_step=1.18s eta=20:26:09 +2026-08-25 20:26:54 | INFO | train rank=0 | step=44710/100000 progress=44.7% loss=0.049316 lr=6.273854e-06 elapsed=16:31:24 avg_step=1.33s recent_step=1.69s eta=20:26:00 +2026-08-25 20:27:06 | INFO | train rank=0 | step=44720/100000 progress=44.7% loss=0.042191 lr=6.272255e-06 elapsed=16:31:36 avg_step=1.33s recent_step=1.14s eta=20:25:45 +2026-08-25 20:27:17 | INFO | train rank=0 | step=44730/100000 progress=44.7% loss=0.042572 lr=6.270656e-06 elapsed=16:31:47 avg_step=1.33s recent_step=1.18s eta=20:25:30 +2026-08-25 20:27:29 | INFO | train rank=0 | step=44740/100000 progress=44.7% loss=0.066589 lr=6.269056e-06 elapsed=16:31:59 avg_step=1.33s recent_step=1.15s eta=20:25:14 +2026-08-25 20:27:43 | INFO | train rank=0 | step=44750/100000 progress=44.8% loss=0.053528 lr=6.267457e-06 elapsed=16:32:13 avg_step=1.33s recent_step=1.40s eta=20:25:02 +2026-08-25 20:27:56 | INFO | train rank=0 | step=44760/100000 progress=44.8% loss=0.046379 lr=6.265857e-06 elapsed=16:32:26 avg_step=1.33s recent_step=1.30s eta=20:24:48 +2026-08-25 20:28:14 | INFO | train rank=0 | step=44770/100000 progress=44.8% loss=0.111816 lr=6.264258e-06 elapsed=16:32:44 avg_step=1.33s recent_step=1.77s eta=20:24:40 +2026-08-25 20:28:26 | INFO | train rank=0 | step=44780/100000 progress=44.8% loss=0.065033 lr=6.262658e-06 elapsed=16:32:56 avg_step=1.33s recent_step=1.22s eta=20:24:25 +2026-08-25 20:28:40 | INFO | train rank=0 | step=44790/100000 progress=44.8% loss=0.062103 lr=6.261058e-06 elapsed=16:33:10 avg_step=1.33s recent_step=1.39s eta=20:24:13 +2026-08-25 20:28:52 | INFO | train rank=0 | step=44800/100000 progress=44.8% loss=0.103149 lr=6.259458e-06 elapsed=16:33:22 avg_step=1.33s recent_step=1.18s eta=20:23:58 +2026-08-25 20:29:05 | INFO | train rank=0 | step=44810/100000 progress=44.8% loss=0.035919 lr=6.257858e-06 elapsed=16:33:35 avg_step=1.33s recent_step=1.31s eta=20:23:44 +2026-08-25 20:29:21 | INFO | train rank=0 | step=44820/100000 progress=44.8% loss=0.039276 lr=6.256257e-06 elapsed=16:33:51 avg_step=1.33s recent_step=1.64s eta=20:23:35 +2026-08-25 20:29:38 | INFO | train rank=0 | step=44830/100000 progress=44.8% loss=0.035019 lr=6.254657e-06 elapsed=16:34:08 avg_step=1.33s recent_step=1.72s eta=20:23:26 +2026-08-25 20:29:50 | INFO | train rank=0 | step=44840/100000 progress=44.8% loss=0.049255 lr=6.253056e-06 elapsed=16:34:20 avg_step=1.33s recent_step=1.14s eta=20:23:11 +2026-08-25 20:30:01 | INFO | train rank=0 | step=44850/100000 progress=44.9% loss=0.050766 lr=6.251455e-06 elapsed=16:34:31 avg_step=1.33s recent_step=1.17s eta=20:22:55 +2026-08-25 20:30:13 | INFO | train rank=0 | step=44860/100000 progress=44.9% loss=0.040146 lr=6.249854e-06 elapsed=16:34:43 avg_step=1.33s recent_step=1.12s eta=20:22:39 +2026-08-25 20:30:27 | INFO | train rank=0 | step=44870/100000 progress=44.9% loss=0.042191 lr=6.248253e-06 elapsed=16:34:56 avg_step=1.33s recent_step=1.39s eta=20:22:27 +2026-08-25 20:30:38 | INFO | train rank=0 | step=44880/100000 progress=44.9% loss=0.074005 lr=6.246652e-06 elapsed=16:35:08 avg_step=1.33s recent_step=1.12s eta=20:22:11 +2026-08-25 20:30:51 | INFO | train rank=0 | step=44890/100000 progress=44.9% loss=0.072968 lr=6.245051e-06 elapsed=16:35:21 avg_step=1.33s recent_step=1.37s eta=20:21:58 +2026-08-25 20:31:06 | INFO | train rank=0 | step=44900/100000 progress=44.9% loss=0.034302 lr=6.243449e-06 elapsed=16:35:35 avg_step=1.33s recent_step=1.41s eta=20:21:46 +2026-08-25 20:31:20 | INFO | train rank=0 | step=44910/100000 progress=44.9% loss=0.035843 lr=6.241848e-06 elapsed=16:35:50 avg_step=1.33s recent_step=1.41s eta=20:21:33 +2026-08-25 20:31:32 | INFO | train rank=0 | step=44920/100000 progress=44.9% loss=0.042068 lr=6.240246e-06 elapsed=16:36:02 avg_step=1.33s recent_step=1.21s eta=20:21:19 +2026-08-25 20:31:46 | INFO | train rank=0 | step=44930/100000 progress=44.9% loss=0.037712 lr=6.238644e-06 elapsed=16:36:16 avg_step=1.33s recent_step=1.45s eta=20:21:07 +2026-08-25 20:32:00 | INFO | train rank=0 | step=44940/100000 progress=44.9% loss=0.045471 lr=6.237042e-06 elapsed=16:36:30 avg_step=1.33s recent_step=1.37s eta=20:20:54 +2026-08-25 20:32:13 | INFO | train rank=0 | step=44950/100000 progress=45.0% loss=0.060684 lr=6.235440e-06 elapsed=16:36:42 avg_step=1.33s recent_step=1.26s eta=20:20:40 +2026-08-25 20:32:26 | INFO | train rank=0 | step=44960/100000 progress=45.0% loss=0.049057 lr=6.233838e-06 elapsed=16:36:56 avg_step=1.33s recent_step=1.31s eta=20:20:26 +2026-08-25 20:32:39 | INFO | train rank=0 | step=44970/100000 progress=45.0% loss=0.043045 lr=6.232235e-06 elapsed=16:37:08 avg_step=1.33s recent_step=1.29s eta=20:20:12 +2026-08-25 20:32:50 | INFO | train rank=0 | step=44980/100000 progress=45.0% loss=0.052704 lr=6.230633e-06 elapsed=16:37:20 avg_step=1.33s recent_step=1.17s eta=20:19:57 +2026-08-25 20:33:05 | INFO | train rank=0 | step=44990/100000 progress=45.0% loss=0.072845 lr=6.229030e-06 elapsed=16:37:35 avg_step=1.33s recent_step=1.49s eta=20:19:46 +2026-08-25 20:33:22 | INFO | train rank=0 | step=45000/100000 progress=45.0% loss=0.054932 lr=6.227427e-06 elapsed=16:37:52 avg_step=1.33s recent_step=1.68s eta=20:19:37 +2026-08-25 20:33:34 | INFO | train rank=0 | step=45010/100000 progress=45.0% loss=0.083008 lr=6.225824e-06 elapsed=16:38:03 avg_step=1.33s recent_step=1.16s eta=20:19:21 +2026-08-25 20:33:48 | INFO | train rank=0 | step=45020/100000 progress=45.0% loss=0.059418 lr=6.224221e-06 elapsed=16:38:18 avg_step=1.33s recent_step=1.44s eta=20:19:09 +2026-08-25 20:34:00 | INFO | train rank=0 | step=45030/100000 progress=45.0% loss=0.056000 lr=6.222618e-06 elapsed=16:38:30 avg_step=1.33s recent_step=1.19s eta=20:18:54 +2026-08-25 20:34:13 | INFO | train rank=0 | step=45040/100000 progress=45.0% loss=0.077576 lr=6.221015e-06 elapsed=16:38:43 avg_step=1.33s recent_step=1.35s eta=20:18:41 +2026-08-25 20:34:25 | INFO | train rank=0 | step=45050/100000 progress=45.1% loss=0.053726 lr=6.219411e-06 elapsed=16:38:55 avg_step=1.33s recent_step=1.15s eta=20:18:26 +2026-08-25 20:34:37 | INFO | train rank=0 | step=45060/100000 progress=45.1% loss=0.033371 lr=6.217808e-06 elapsed=16:39:07 avg_step=1.33s recent_step=1.25s eta=20:18:11 +2026-08-25 20:34:49 | INFO | train rank=0 | step=45070/100000 progress=45.1% loss=0.065887 lr=6.216204e-06 elapsed=16:39:19 avg_step=1.33s recent_step=1.16s eta=20:17:56 +2026-08-25 20:35:01 | INFO | train rank=0 | step=45080/100000 progress=45.1% loss=0.071915 lr=6.214600e-06 elapsed=16:39:31 avg_step=1.33s recent_step=1.20s eta=20:17:41 +2026-08-25 20:35:13 | INFO | train rank=0 | step=45090/100000 progress=45.1% loss=0.045959 lr=6.212996e-06 elapsed=16:39:43 avg_step=1.33s recent_step=1.25s eta=20:17:27 +2026-08-25 20:35:27 | INFO | train rank=0 | step=45100/100000 progress=45.1% loss=0.047440 lr=6.211392e-06 elapsed=16:39:57 avg_step=1.33s recent_step=1.41s eta=20:17:15 +2026-08-25 20:35:41 | INFO | train rank=0 | step=45110/100000 progress=45.1% loss=0.049301 lr=6.209788e-06 elapsed=16:40:11 avg_step=1.33s recent_step=1.32s eta=20:17:01 +2026-08-25 20:35:53 | INFO | train rank=0 | step=45120/100000 progress=45.1% loss=0.086426 lr=6.208183e-06 elapsed=16:40:23 avg_step=1.33s recent_step=1.22s eta=20:16:46 +2026-08-25 20:36:09 | INFO | train rank=0 | step=45130/100000 progress=45.1% loss=0.066818 lr=6.206579e-06 elapsed=16:40:39 avg_step=1.33s recent_step=1.60s eta=20:16:36 +2026-08-25 20:36:25 | INFO | train rank=0 | step=45140/100000 progress=45.1% loss=0.034988 lr=6.204974e-06 elapsed=16:40:55 avg_step=1.33s recent_step=1.61s eta=20:16:27 +2026-08-25 20:36:38 | INFO | train rank=0 | step=45150/100000 progress=45.1% loss=0.030029 lr=6.203369e-06 elapsed=16:41:07 avg_step=1.33s recent_step=1.26s eta=20:16:12 +2026-08-25 20:36:51 | INFO | train rank=0 | step=45160/100000 progress=45.2% loss=0.041092 lr=6.201764e-06 elapsed=16:41:21 avg_step=1.33s recent_step=1.36s eta=20:15:59 +2026-08-25 20:37:06 | INFO | train rank=0 | step=45170/100000 progress=45.2% loss=0.044991 lr=6.200159e-06 elapsed=16:41:36 avg_step=1.33s recent_step=1.51s eta=20:15:48 +2026-08-25 20:37:18 | INFO | train rank=0 | step=45180/100000 progress=45.2% loss=0.082184 lr=6.198554e-06 elapsed=16:41:48 avg_step=1.33s recent_step=1.20s eta=20:15:33 +2026-08-25 20:37:35 | INFO | train rank=0 | step=45190/100000 progress=45.2% loss=0.040359 lr=6.196949e-06 elapsed=16:42:05 avg_step=1.33s recent_step=1.66s eta=20:15:24 +2026-08-25 20:37:46 | INFO | train rank=0 | step=45200/100000 progress=45.2% loss=0.051636 lr=6.195343e-06 elapsed=16:42:16 avg_step=1.33s recent_step=1.15s eta=20:15:09 +2026-08-25 20:37:58 | INFO | train rank=0 | step=45210/100000 progress=45.2% loss=0.070892 lr=6.193738e-06 elapsed=16:42:27 avg_step=1.33s recent_step=1.12s eta=20:14:53 +2026-08-25 20:38:13 | INFO | train rank=0 | step=45220/100000 progress=45.2% loss=0.039574 lr=6.192132e-06 elapsed=16:42:42 avg_step=1.33s recent_step=1.51s eta=20:14:42 +2026-08-25 20:38:29 | INFO | train rank=0 | step=45230/100000 progress=45.2% loss=0.053635 lr=6.190526e-06 elapsed=16:42:59 avg_step=1.33s recent_step=1.63s eta=20:14:32 +2026-08-25 20:38:41 | INFO | train rank=0 | step=45240/100000 progress=45.2% loss=0.058151 lr=6.188920e-06 elapsed=16:43:11 avg_step=1.33s recent_step=1.18s eta=20:14:17 +2026-08-25 20:38:53 | INFO | train rank=0 | step=45250/100000 progress=45.2% loss=0.069946 lr=6.187314e-06 elapsed=16:43:23 avg_step=1.33s recent_step=1.20s eta=20:14:02 +2026-08-25 20:39:07 | INFO | train rank=0 | step=45260/100000 progress=45.3% loss=0.060516 lr=6.185708e-06 elapsed=16:43:37 avg_step=1.33s recent_step=1.39s eta=20:13:49 +2026-08-25 20:39:21 | INFO | train rank=0 | step=45270/100000 progress=45.3% loss=0.043701 lr=6.184101e-06 elapsed=16:43:51 avg_step=1.33s recent_step=1.43s eta=20:13:37 +2026-08-25 20:39:33 | INFO | train rank=0 | step=45280/100000 progress=45.3% loss=0.047302 lr=6.182495e-06 elapsed=16:44:03 avg_step=1.33s recent_step=1.22s eta=20:13:23 +2026-08-25 20:39:45 | INFO | train rank=0 | step=45290/100000 progress=45.3% loss=0.048508 lr=6.180888e-06 elapsed=16:44:15 avg_step=1.33s recent_step=1.19s eta=20:13:08 +2026-08-25 20:39:56 | INFO | train rank=0 | step=45300/100000 progress=45.3% loss=0.047623 lr=6.179282e-06 elapsed=16:44:26 avg_step=1.33s recent_step=1.13s eta=20:12:52 +2026-08-25 20:40:08 | INFO | train rank=0 | step=45310/100000 progress=45.3% loss=0.047836 lr=6.177675e-06 elapsed=16:44:38 avg_step=1.33s recent_step=1.16s eta=20:12:36 +2026-08-25 20:40:22 | INFO | train rank=0 | step=45320/100000 progress=45.3% loss=0.058685 lr=6.176068e-06 elapsed=16:44:52 avg_step=1.33s recent_step=1.40s eta=20:12:24 +2026-08-25 20:40:34 | INFO | train rank=0 | step=45330/100000 progress=45.3% loss=0.060791 lr=6.174461e-06 elapsed=16:45:04 avg_step=1.33s recent_step=1.25s eta=20:12:10 +2026-08-25 20:40:51 | INFO | train rank=0 | step=45340/100000 progress=45.3% loss=0.041351 lr=6.172853e-06 elapsed=16:45:21 avg_step=1.33s recent_step=1.69s eta=20:12:01 +2026-08-25 20:41:08 | INFO | train rank=0 | step=45350/100000 progress=45.4% loss=0.043671 lr=6.171246e-06 elapsed=16:45:38 avg_step=1.33s recent_step=1.65s eta=20:11:51 +2026-08-25 20:41:24 | INFO | train rank=0 | step=45360/100000 progress=45.4% loss=0.086975 lr=6.169638e-06 elapsed=16:45:54 avg_step=1.33s recent_step=1.58s eta=20:11:41 +2026-08-25 20:41:39 | INFO | train rank=0 | step=45370/100000 progress=45.4% loss=0.052170 lr=6.168031e-06 elapsed=16:46:09 avg_step=1.33s recent_step=1.58s eta=20:11:31 +2026-08-25 20:41:51 | INFO | train rank=0 | step=45380/100000 progress=45.4% loss=0.045883 lr=6.166423e-06 elapsed=16:46:21 avg_step=1.33s recent_step=1.15s eta=20:11:15 +2026-08-25 20:42:03 | INFO | train rank=0 | step=45390/100000 progress=45.4% loss=0.045036 lr=6.164815e-06 elapsed=16:46:33 avg_step=1.33s recent_step=1.23s eta=20:11:01 +2026-08-25 20:42:15 | INFO | train rank=0 | step=45400/100000 progress=45.4% loss=0.034943 lr=6.163207e-06 elapsed=16:46:45 avg_step=1.33s recent_step=1.15s eta=20:10:45 +2026-08-25 20:42:27 | INFO | train rank=0 | step=45410/100000 progress=45.4% loss=0.056900 lr=6.161599e-06 elapsed=16:46:56 avg_step=1.33s recent_step=1.17s eta=20:10:30 +2026-08-25 20:42:38 | INFO | train rank=0 | step=45420/100000 progress=45.4% loss=0.063477 lr=6.159991e-06 elapsed=16:47:08 avg_step=1.33s recent_step=1.18s eta=20:10:15 +2026-08-25 20:42:58 | INFO | train rank=0 | step=45430/100000 progress=45.4% loss=0.039932 lr=6.158382e-06 elapsed=16:47:28 avg_step=1.33s recent_step=1.95s eta=20:10:09 +2026-08-25 20:43:11 | INFO | train rank=0 | step=45440/100000 progress=45.4% loss=0.047211 lr=6.156774e-06 elapsed=16:47:41 avg_step=1.33s recent_step=1.28s eta=20:09:55 +2026-08-25 20:43:23 | INFO | train rank=0 | step=45450/100000 progress=45.5% loss=0.048401 lr=6.155165e-06 elapsed=16:47:53 avg_step=1.33s recent_step=1.24s eta=20:09:41 +2026-08-25 20:43:37 | INFO | train rank=0 | step=45460/100000 progress=45.5% loss=0.055923 lr=6.153556e-06 elapsed=16:48:06 avg_step=1.33s recent_step=1.35s eta=20:09:28 +2026-08-25 20:43:50 | INFO | train rank=0 | step=45470/100000 progress=45.5% loss=0.049561 lr=6.151947e-06 elapsed=16:48:20 avg_step=1.33s recent_step=1.34s eta=20:09:15 +2026-08-25 20:44:02 | INFO | train rank=0 | step=45480/100000 progress=45.5% loss=0.054230 lr=6.150338e-06 elapsed=16:48:32 avg_step=1.33s recent_step=1.22s eta=20:09:00 +2026-08-25 20:44:21 | INFO | train rank=0 | step=45490/100000 progress=45.5% loss=0.033478 lr=6.148729e-06 elapsed=16:48:51 avg_step=1.33s recent_step=1.89s eta=20:08:53 +2026-08-25 20:44:34 | INFO | train rank=0 | step=45500/100000 progress=45.5% loss=0.041183 lr=6.147120e-06 elapsed=16:49:04 avg_step=1.33s recent_step=1.26s eta=20:08:39 +2026-08-25 20:44:45 | INFO | train rank=0 | step=45510/100000 progress=45.5% loss=0.033966 lr=6.145510e-06 elapsed=16:49:15 avg_step=1.33s recent_step=1.14s eta=20:08:24 +2026-08-25 20:44:57 | INFO | train rank=0 | step=45520/100000 progress=45.5% loss=0.057770 lr=6.143901e-06 elapsed=16:49:27 avg_step=1.33s recent_step=1.18s eta=20:08:09 +2026-08-25 20:45:09 | INFO | train rank=0 | step=45530/100000 progress=45.5% loss=0.059586 lr=6.142291e-06 elapsed=16:49:39 avg_step=1.33s recent_step=1.20s eta=20:07:54 +2026-08-25 20:45:22 | INFO | train rank=0 | step=45540/100000 progress=45.5% loss=0.052689 lr=6.140681e-06 elapsed=16:49:52 avg_step=1.33s recent_step=1.28s eta=20:07:40 +2026-08-25 20:45:36 | INFO | train rank=0 | step=45550/100000 progress=45.6% loss=0.057220 lr=6.139071e-06 elapsed=16:50:06 avg_step=1.33s recent_step=1.46s eta=20:07:28 +2026-08-25 20:45:48 | INFO | train rank=0 | step=45560/100000 progress=45.6% loss=0.037186 lr=6.137461e-06 elapsed=16:50:18 avg_step=1.33s recent_step=1.14s eta=20:07:12 +2026-08-25 20:46:01 | INFO | train rank=0 | step=45570/100000 progress=45.6% loss=0.082703 lr=6.135851e-06 elapsed=16:50:31 avg_step=1.33s recent_step=1.34s eta=20:06:59 +2026-08-25 20:46:15 | INFO | train rank=0 | step=45580/100000 progress=45.6% loss=0.043594 lr=6.134241e-06 elapsed=16:50:45 avg_step=1.33s recent_step=1.36s eta=20:06:46 +2026-08-25 20:46:30 | INFO | train rank=0 | step=45590/100000 progress=45.6% loss=0.039009 lr=6.132630e-06 elapsed=16:51:00 avg_step=1.33s recent_step=1.49s eta=20:06:35 +2026-08-25 20:46:45 | INFO | train rank=0 | step=45600/100000 progress=45.6% loss=0.043991 lr=6.131020e-06 elapsed=16:51:15 avg_step=1.33s recent_step=1.53s eta=20:06:24 +2026-08-25 20:46:57 | INFO | train rank=0 | step=45610/100000 progress=45.6% loss=0.076202 lr=6.129409e-06 elapsed=16:51:27 avg_step=1.33s recent_step=1.18s eta=20:06:09 +2026-08-25 20:47:14 | INFO | train rank=0 | step=45620/100000 progress=45.6% loss=0.062820 lr=6.127798e-06 elapsed=16:51:44 avg_step=1.33s recent_step=1.76s eta=20:06:01 +2026-08-25 20:47:27 | INFO | train rank=0 | step=45630/100000 progress=45.6% loss=0.049149 lr=6.126187e-06 elapsed=16:51:57 avg_step=1.33s recent_step=1.30s eta=20:05:47 +2026-08-25 20:47:42 | INFO | train rank=0 | step=45640/100000 progress=45.6% loss=0.043259 lr=6.124576e-06 elapsed=16:52:12 avg_step=1.33s recent_step=1.43s eta=20:05:35 +2026-08-25 20:47:53 | INFO | train rank=0 | step=45650/100000 progress=45.6% loss=0.030533 lr=6.122965e-06 elapsed=16:52:23 avg_step=1.33s recent_step=1.15s eta=20:05:20 +2026-08-25 20:48:05 | INFO | train rank=0 | step=45660/100000 progress=45.7% loss=0.123169 lr=6.121354e-06 elapsed=16:52:35 avg_step=1.33s recent_step=1.20s eta=20:05:05 +2026-08-25 20:48:16 | INFO | train rank=0 | step=45670/100000 progress=45.7% loss=0.050186 lr=6.119742e-06 elapsed=16:52:46 avg_step=1.33s recent_step=1.10s eta=20:04:49 +2026-08-25 20:48:28 | INFO | train rank=0 | step=45680/100000 progress=45.7% loss=0.039337 lr=6.118131e-06 elapsed=16:52:58 avg_step=1.33s recent_step=1.20s eta=20:04:34 +2026-08-25 20:48:43 | INFO | train rank=0 | step=45690/100000 progress=45.7% loss=0.060501 lr=6.116519e-06 elapsed=16:53:13 avg_step=1.33s recent_step=1.44s eta=20:04:22 +2026-08-25 20:48:59 | INFO | train rank=0 | step=45700/100000 progress=45.7% loss=0.044846 lr=6.114907e-06 elapsed=16:53:29 avg_step=1.33s recent_step=1.64s eta=20:04:12 +2026-08-25 20:49:13 | INFO | train rank=0 | step=45710/100000 progress=45.7% loss=0.041107 lr=6.113296e-06 elapsed=16:53:43 avg_step=1.33s recent_step=1.38s eta=20:04:00 +2026-08-25 20:49:27 | INFO | train rank=0 | step=45720/100000 progress=45.7% loss=0.058716 lr=6.111683e-06 elapsed=16:53:57 avg_step=1.33s recent_step=1.43s eta=20:03:48 +2026-08-25 20:49:40 | INFO | train rank=0 | step=45730/100000 progress=45.7% loss=0.054123 lr=6.110071e-06 elapsed=16:54:10 avg_step=1.33s recent_step=1.29s eta=20:03:34 +2026-08-25 20:49:53 | INFO | train rank=0 | step=45740/100000 progress=45.7% loss=0.068787 lr=6.108459e-06 elapsed=16:54:23 avg_step=1.33s recent_step=1.28s eta=20:03:20 +2026-08-25 20:50:06 | INFO | train rank=0 | step=45750/100000 progress=45.8% loss=0.033157 lr=6.106847e-06 elapsed=16:54:35 avg_step=1.33s recent_step=1.26s eta=20:03:06 +2026-08-25 20:50:17 | INFO | train rank=0 | step=45760/100000 progress=45.8% loss=0.041626 lr=6.105234e-06 elapsed=16:54:47 avg_step=1.33s recent_step=1.17s eta=20:02:51 +2026-08-25 20:50:35 | INFO | train rank=0 | step=45770/100000 progress=45.8% loss=0.056778 lr=6.103622e-06 elapsed=16:55:05 avg_step=1.33s recent_step=1.79s eta=20:02:43 +2026-08-25 20:50:48 | INFO | train rank=0 | step=45780/100000 progress=45.8% loss=0.042122 lr=6.102009e-06 elapsed=16:55:18 avg_step=1.33s recent_step=1.30s eta=20:02:29 +2026-08-25 20:51:02 | INFO | train rank=0 | step=45790/100000 progress=45.8% loss=0.036209 lr=6.100396e-06 elapsed=16:55:32 avg_step=1.33s recent_step=1.38s eta=20:02:16 +2026-08-25 20:51:15 | INFO | train rank=0 | step=45800/100000 progress=45.8% loss=0.047379 lr=6.098783e-06 elapsed=16:55:45 avg_step=1.33s recent_step=1.31s eta=20:02:03 +2026-08-25 20:51:27 | INFO | train rank=0 | step=45810/100000 progress=45.8% loss=0.040283 lr=6.097170e-06 elapsed=16:55:57 avg_step=1.33s recent_step=1.19s eta=20:01:48 +2026-08-25 20:51:38 | INFO | train rank=0 | step=45820/100000 progress=45.8% loss=0.064026 lr=6.095557e-06 elapsed=16:56:08 avg_step=1.33s recent_step=1.14s eta=20:01:32 +2026-08-25 20:51:51 | INFO | train rank=0 | step=45830/100000 progress=45.8% loss=0.035896 lr=6.093943e-06 elapsed=16:56:21 avg_step=1.33s recent_step=1.29s eta=20:01:18 +2026-08-25 20:52:05 | INFO | train rank=0 | step=45840/100000 progress=45.8% loss=0.067963 lr=6.092330e-06 elapsed=16:56:35 avg_step=1.33s recent_step=1.36s eta=20:01:05 +2026-08-25 20:52:16 | INFO | train rank=0 | step=45850/100000 progress=45.9% loss=0.045197 lr=6.090716e-06 elapsed=16:56:46 avg_step=1.33s recent_step=1.16s eta=20:00:50 +2026-08-25 20:52:32 | INFO | train rank=0 | step=45860/100000 progress=45.9% loss=0.058762 lr=6.089102e-06 elapsed=16:57:02 avg_step=1.33s recent_step=1.56s eta=20:00:40 +2026-08-25 20:52:44 | INFO | train rank=0 | step=45870/100000 progress=45.9% loss=0.034203 lr=6.087489e-06 elapsed=16:57:14 avg_step=1.33s recent_step=1.17s eta=20:00:24 +2026-08-25 20:53:01 | INFO | train rank=0 | step=45880/100000 progress=45.9% loss=0.036499 lr=6.085875e-06 elapsed=16:57:31 avg_step=1.33s recent_step=1.75s eta=20:00:16 +2026-08-25 20:53:13 | INFO | train rank=0 | step=45890/100000 progress=45.9% loss=0.066986 lr=6.084261e-06 elapsed=16:57:43 avg_step=1.33s recent_step=1.16s eta=20:00:01 +2026-08-25 20:53:32 | INFO | train rank=0 | step=45900/100000 progress=45.9% loss=0.042191 lr=6.082646e-06 elapsed=16:58:02 avg_step=1.33s recent_step=1.88s eta=19:59:54 +2026-08-25 20:53:45 | INFO | train rank=0 | step=45910/100000 progress=45.9% loss=0.051895 lr=6.081032e-06 elapsed=16:58:14 avg_step=1.33s recent_step=1.28s eta=19:59:40 +2026-08-25 20:53:56 | INFO | train rank=0 | step=45920/100000 progress=45.9% loss=0.026443 lr=6.079418e-06 elapsed=16:58:26 avg_step=1.33s recent_step=1.16s eta=19:59:25 +2026-08-25 20:54:07 | INFO | train rank=0 | step=45930/100000 progress=45.9% loss=0.041077 lr=6.077803e-06 elapsed=16:58:37 avg_step=1.33s recent_step=1.13s eta=19:59:09 +2026-08-25 20:54:21 | INFO | train rank=0 | step=45940/100000 progress=45.9% loss=0.053116 lr=6.076189e-06 elapsed=16:58:51 avg_step=1.33s recent_step=1.37s eta=19:58:56 +2026-08-25 20:54:35 | INFO | train rank=0 | step=45950/100000 progress=46.0% loss=0.067352 lr=6.074574e-06 elapsed=16:59:05 avg_step=1.33s recent_step=1.44s eta=19:58:44 +2026-08-25 20:54:50 | INFO | train rank=0 | step=45960/100000 progress=46.0% loss=0.038132 lr=6.072959e-06 elapsed=16:59:19 avg_step=1.33s recent_step=1.40s eta=19:58:32 +2026-08-25 20:55:02 | INFO | train rank=0 | step=45970/100000 progress=46.0% loss=0.045624 lr=6.071344e-06 elapsed=16:59:32 avg_step=1.33s recent_step=1.23s eta=19:58:17 +2026-08-25 20:55:19 | INFO | train rank=0 | step=45980/100000 progress=46.0% loss=0.027458 lr=6.069729e-06 elapsed=16:59:49 avg_step=1.33s recent_step=1.73s eta=19:58:08 +2026-08-25 20:55:30 | INFO | train rank=0 | step=45990/100000 progress=46.0% loss=0.054520 lr=6.068114e-06 elapsed=17:00:00 avg_step=1.33s recent_step=1.13s eta=19:57:53 +2026-08-25 20:55:42 | INFO | train rank=0 | step=46000/100000 progress=46.0% loss=0.043396 lr=6.066498e-06 elapsed=17:00:12 avg_step=1.33s recent_step=1.17s eta=19:57:38 +2026-08-25 20:55:57 | INFO | train rank=0 | step=46010/100000 progress=46.0% loss=0.045425 lr=6.064883e-06 elapsed=17:00:27 avg_step=1.33s recent_step=1.48s eta=19:57:26 +2026-08-25 20:56:13 | INFO | train rank=0 | step=46020/100000 progress=46.0% loss=0.069672 lr=6.063267e-06 elapsed=17:00:43 avg_step=1.33s recent_step=1.62s eta=19:57:16 +2026-08-25 20:56:28 | INFO | train rank=0 | step=46030/100000 progress=46.0% loss=0.049561 lr=6.061651e-06 elapsed=17:00:58 avg_step=1.33s recent_step=1.47s eta=19:57:05 +2026-08-25 20:56:40 | INFO | train rank=0 | step=46040/100000 progress=46.0% loss=0.054138 lr=6.060036e-06 elapsed=17:01:10 avg_step=1.33s recent_step=1.18s eta=19:56:49 +2026-08-25 20:56:51 | INFO | train rank=0 | step=46050/100000 progress=46.1% loss=0.040436 lr=6.058420e-06 elapsed=17:01:21 avg_step=1.33s recent_step=1.16s eta=19:56:34 +2026-08-25 20:57:05 | INFO | train rank=0 | step=46060/100000 progress=46.1% loss=0.034592 lr=6.056804e-06 elapsed=17:01:35 avg_step=1.33s recent_step=1.40s eta=19:56:22 +2026-08-25 20:57:20 | INFO | train rank=0 | step=46070/100000 progress=46.1% loss=0.047882 lr=6.055187e-06 elapsed=17:01:50 avg_step=1.33s recent_step=1.44s eta=19:56:10 +2026-08-25 20:57:32 | INFO | train rank=0 | step=46080/100000 progress=46.1% loss=0.075256 lr=6.053571e-06 elapsed=17:02:02 avg_step=1.33s recent_step=1.20s eta=19:55:55 +2026-08-25 20:57:44 | INFO | train rank=0 | step=46090/100000 progress=46.1% loss=0.048096 lr=6.051955e-06 elapsed=17:02:14 avg_step=1.33s recent_step=1.24s eta=19:55:40 +2026-08-25 20:57:56 | INFO | train rank=0 | step=46100/100000 progress=46.1% loss=0.043640 lr=6.050338e-06 elapsed=17:02:25 avg_step=1.33s recent_step=1.15s eta=19:55:25 +2026-08-25 20:58:11 | INFO | train rank=0 | step=46110/100000 progress=46.1% loss=0.039215 lr=6.048722e-06 elapsed=17:02:41 avg_step=1.33s recent_step=1.57s eta=19:55:15 +2026-08-25 20:58:29 | INFO | train rank=0 | step=46120/100000 progress=46.1% loss=0.064941 lr=6.047105e-06 elapsed=17:02:59 avg_step=1.33s recent_step=1.79s eta=19:55:07 +2026-08-25 20:58:41 | INFO | train rank=0 | step=46130/100000 progress=46.1% loss=0.054428 lr=6.045488e-06 elapsed=17:03:11 avg_step=1.33s recent_step=1.15s eta=19:54:51 +2026-08-25 20:58:52 | INFO | train rank=0 | step=46140/100000 progress=46.1% loss=0.076996 lr=6.043871e-06 elapsed=17:03:22 avg_step=1.33s recent_step=1.17s eta=19:54:36 +2026-08-25 20:59:04 | INFO | train rank=0 | step=46150/100000 progress=46.2% loss=0.048615 lr=6.042254e-06 elapsed=17:03:34 avg_step=1.33s recent_step=1.15s eta=19:54:21 +2026-08-25 20:59:15 | INFO | train rank=0 | step=46160/100000 progress=46.2% loss=0.056084 lr=6.040637e-06 elapsed=17:03:45 avg_step=1.33s recent_step=1.13s eta=19:54:05 +2026-08-25 20:59:30 | INFO | train rank=0 | step=46170/100000 progress=46.2% loss=0.063751 lr=6.039019e-06 elapsed=17:04:00 avg_step=1.33s recent_step=1.45s eta=19:53:53 +2026-08-25 20:59:45 | INFO | train rank=0 | step=46180/100000 progress=46.2% loss=0.085876 lr=6.037402e-06 elapsed=17:04:15 avg_step=1.33s recent_step=1.56s eta=19:53:43 +2026-08-25 20:59:57 | INFO | train rank=0 | step=46190/100000 progress=46.2% loss=0.033577 lr=6.035784e-06 elapsed=17:04:27 avg_step=1.33s recent_step=1.16s eta=19:53:27 +2026-08-25 21:00:10 | INFO | train rank=0 | step=46200/100000 progress=46.2% loss=0.053101 lr=6.034167e-06 elapsed=17:04:40 avg_step=1.33s recent_step=1.34s eta=19:53:14 +2026-08-25 21:00:26 | INFO | train rank=0 | step=46210/100000 progress=46.2% loss=0.050293 lr=6.032549e-06 elapsed=17:04:56 avg_step=1.33s recent_step=1.55s eta=19:53:03 +2026-08-25 21:00:38 | INFO | train rank=0 | step=46220/100000 progress=46.2% loss=0.048828 lr=6.030931e-06 elapsed=17:05:08 avg_step=1.33s recent_step=1.18s eta=19:52:48 +2026-08-25 21:00:49 | INFO | train rank=0 | step=46230/100000 progress=46.2% loss=0.050415 lr=6.029313e-06 elapsed=17:05:19 avg_step=1.33s recent_step=1.18s eta=19:52:33 +2026-08-25 21:01:02 | INFO | train rank=0 | step=46240/100000 progress=46.2% loss=0.061554 lr=6.027695e-06 elapsed=17:05:32 avg_step=1.33s recent_step=1.22s eta=19:52:19 +2026-08-25 21:01:17 | INFO | train rank=0 | step=46250/100000 progress=46.2% loss=0.039520 lr=6.026077e-06 elapsed=17:05:47 avg_step=1.33s recent_step=1.52s eta=19:52:07 +2026-08-25 21:01:29 | INFO | train rank=0 | step=46260/100000 progress=46.3% loss=0.037506 lr=6.024458e-06 elapsed=17:05:59 avg_step=1.33s recent_step=1.25s eta=19:51:53 +2026-08-25 21:01:41 | INFO | train rank=0 | step=46270/100000 progress=46.3% loss=0.051453 lr=6.022840e-06 elapsed=17:06:10 avg_step=1.33s recent_step=1.12s eta=19:51:37 +2026-08-25 21:01:52 | INFO | train rank=0 | step=46280/100000 progress=46.3% loss=0.041672 lr=6.021221e-06 elapsed=17:06:22 avg_step=1.33s recent_step=1.18s eta=19:51:22 +2026-08-25 21:02:04 | INFO | train rank=0 | step=46290/100000 progress=46.3% loss=0.065552 lr=6.019603e-06 elapsed=17:06:34 avg_step=1.33s recent_step=1.16s eta=19:51:07 +2026-08-25 21:02:16 | INFO | train rank=0 | step=46300/100000 progress=46.3% loss=0.082886 lr=6.017984e-06 elapsed=17:06:45 avg_step=1.33s recent_step=1.16s eta=19:50:52 +2026-08-25 21:02:31 | INFO | train rank=0 | step=46310/100000 progress=46.3% loss=0.048920 lr=6.016365e-06 elapsed=17:07:01 avg_step=1.33s recent_step=1.56s eta=19:50:41 +2026-08-25 21:02:46 | INFO | train rank=0 | step=46320/100000 progress=46.3% loss=0.061890 lr=6.014746e-06 elapsed=17:07:15 avg_step=1.33s recent_step=1.43s eta=19:50:29 +2026-08-25 21:02:59 | INFO | train rank=0 | step=46330/100000 progress=46.3% loss=0.035858 lr=6.013127e-06 elapsed=17:07:29 avg_step=1.33s recent_step=1.37s eta=19:50:16 +2026-08-25 21:03:11 | INFO | train rank=0 | step=46340/100000 progress=46.3% loss=0.042770 lr=6.011508e-06 elapsed=17:07:41 avg_step=1.33s recent_step=1.17s eta=19:50:01 +2026-08-25 21:03:27 | INFO | train rank=0 | step=46350/100000 progress=46.4% loss=0.048340 lr=6.009888e-06 elapsed=17:07:57 avg_step=1.33s recent_step=1.60s eta=19:49:51 +2026-08-25 21:03:42 | INFO | train rank=0 | step=46360/100000 progress=46.4% loss=0.052383 lr=6.008269e-06 elapsed=17:08:12 avg_step=1.33s recent_step=1.53s eta=19:49:40 +2026-08-25 21:03:55 | INFO | train rank=0 | step=46370/100000 progress=46.4% loss=0.042755 lr=6.006649e-06 elapsed=17:08:25 avg_step=1.33s recent_step=1.28s eta=19:49:26 +2026-08-25 21:04:07 | INFO | train rank=0 | step=46380/100000 progress=46.4% loss=0.075745 lr=6.005030e-06 elapsed=17:08:37 avg_step=1.33s recent_step=1.19s eta=19:49:11 +2026-08-25 21:04:21 | INFO | train rank=0 | step=46390/100000 progress=46.4% loss=0.104736 lr=6.003410e-06 elapsed=17:08:51 avg_step=1.33s recent_step=1.41s eta=19:48:58 +2026-08-25 21:04:38 | INFO | train rank=0 | step=46400/100000 progress=46.4% loss=0.067108 lr=6.001790e-06 elapsed=17:09:08 avg_step=1.33s recent_step=1.76s eta=19:48:50 +2026-08-25 21:04:50 | INFO | train rank=0 | step=46410/100000 progress=46.4% loss=0.061340 lr=6.000170e-06 elapsed=17:09:20 avg_step=1.33s recent_step=1.16s eta=19:48:35 +2026-08-25 21:05:03 | INFO | train rank=0 | step=46420/100000 progress=46.4% loss=0.048141 lr=5.998550e-06 elapsed=17:09:33 avg_step=1.33s recent_step=1.34s eta=19:48:22 +2026-08-25 21:05:15 | INFO | train rank=0 | step=46430/100000 progress=46.4% loss=0.045563 lr=5.996930e-06 elapsed=17:09:45 avg_step=1.33s recent_step=1.18s eta=19:48:06 +2026-08-25 21:05:31 | INFO | train rank=0 | step=46440/100000 progress=46.4% loss=0.072144 lr=5.995309e-06 elapsed=17:10:01 avg_step=1.33s recent_step=1.55s eta=19:47:56 +2026-08-25 21:05:45 | INFO | train rank=0 | step=46450/100000 progress=46.5% loss=0.048248 lr=5.993689e-06 elapsed=17:10:15 avg_step=1.33s recent_step=1.40s eta=19:47:43 +2026-08-25 21:05:56 | INFO | train rank=0 | step=46460/100000 progress=46.5% loss=0.061829 lr=5.992068e-06 elapsed=17:10:26 avg_step=1.33s recent_step=1.13s eta=19:47:28 +2026-08-25 21:06:11 | INFO | train rank=0 | step=46470/100000 progress=46.5% loss=0.041656 lr=5.990448e-06 elapsed=17:10:41 avg_step=1.33s recent_step=1.53s eta=19:47:17 +2026-08-25 21:06:25 | INFO | train rank=0 | step=46480/100000 progress=46.5% loss=0.042862 lr=5.988827e-06 elapsed=17:10:55 avg_step=1.33s recent_step=1.37s eta=19:47:04 +2026-08-25 21:06:37 | INFO | train rank=0 | step=46490/100000 progress=46.5% loss=0.105896 lr=5.987206e-06 elapsed=17:11:07 avg_step=1.33s recent_step=1.17s eta=19:46:49 +2026-08-25 21:06:51 | INFO | train rank=0 | step=46500/100000 progress=46.5% loss=0.101013 lr=5.985585e-06 elapsed=17:11:21 avg_step=1.33s recent_step=1.44s eta=19:46:36 +2026-08-25 21:07:06 | INFO | train rank=0 | step=46510/100000 progress=46.5% loss=0.046448 lr=5.983964e-06 elapsed=17:11:35 avg_step=1.33s recent_step=1.45s eta=19:46:25 +2026-08-25 21:07:23 | INFO | train rank=0 | step=46520/100000 progress=46.5% loss=0.070435 lr=5.982343e-06 elapsed=17:11:53 avg_step=1.33s recent_step=1.74s eta=19:46:16 +2026-08-25 21:07:35 | INFO | train rank=0 | step=46530/100000 progress=46.5% loss=0.051590 lr=5.980722e-06 elapsed=17:12:05 avg_step=1.33s recent_step=1.18s eta=19:46:01 +2026-08-25 21:07:46 | INFO | train rank=0 | step=46540/100000 progress=46.5% loss=0.068634 lr=5.979100e-06 elapsed=17:12:16 avg_step=1.33s recent_step=1.13s eta=19:45:45 +2026-08-25 21:07:58 | INFO | train rank=0 | step=46550/100000 progress=46.6% loss=0.068665 lr=5.977479e-06 elapsed=17:12:27 avg_step=1.33s recent_step=1.14s eta=19:45:30 +2026-08-25 21:08:11 | INFO | train rank=0 | step=46560/100000 progress=46.6% loss=0.065033 lr=5.975857e-06 elapsed=17:12:41 avg_step=1.33s recent_step=1.35s eta=19:45:17 +2026-08-25 21:08:23 | INFO | train rank=0 | step=46570/100000 progress=46.6% loss=0.083694 lr=5.974235e-06 elapsed=17:12:53 avg_step=1.33s recent_step=1.21s eta=19:45:02 +2026-08-25 21:08:35 | INFO | train rank=0 | step=46580/100000 progress=46.6% loss=0.060974 lr=5.972613e-06 elapsed=17:13:05 avg_step=1.33s recent_step=1.18s eta=19:44:47 +2026-08-25 21:08:49 | INFO | train rank=0 | step=46590/100000 progress=46.6% loss=0.050964 lr=5.970992e-06 elapsed=17:13:19 avg_step=1.33s recent_step=1.42s eta=19:44:35 +2026-08-25 21:09:01 | INFO | train rank=0 | step=46600/100000 progress=46.6% loss=0.047310 lr=5.969369e-06 elapsed=17:13:31 avg_step=1.33s recent_step=1.16s eta=19:44:19 +2026-08-25 21:09:12 | INFO | train rank=0 | step=46610/100000 progress=46.6% loss=0.026833 lr=5.967747e-06 elapsed=17:13:42 avg_step=1.33s recent_step=1.15s eta=19:44:04 +2026-08-25 21:09:26 | INFO | train rank=0 | step=46620/100000 progress=46.6% loss=0.038315 lr=5.966125e-06 elapsed=17:13:56 avg_step=1.33s recent_step=1.36s eta=19:43:51 +2026-08-25 21:09:37 | INFO | train rank=0 | step=46630/100000 progress=46.6% loss=0.051987 lr=5.964503e-06 elapsed=17:14:07 avg_step=1.33s recent_step=1.15s eta=19:43:36 +2026-08-25 21:09:49 | INFO | train rank=0 | step=46640/100000 progress=46.6% loss=0.042511 lr=5.962880e-06 elapsed=17:14:19 avg_step=1.33s recent_step=1.16s eta=19:43:21 +2026-08-25 21:10:01 | INFO | train rank=0 | step=46650/100000 progress=46.7% loss=0.058044 lr=5.961258e-06 elapsed=17:14:31 avg_step=1.33s recent_step=1.19s eta=19:43:06 +2026-08-25 21:10:19 | INFO | train rank=0 | step=46660/100000 progress=46.7% loss=0.054169 lr=5.959635e-06 elapsed=17:14:49 avg_step=1.33s recent_step=1.80s eta=19:42:58 +2026-08-25 21:10:34 | INFO | train rank=0 | step=46670/100000 progress=46.7% loss=0.067902 lr=5.958012e-06 elapsed=17:15:04 avg_step=1.33s recent_step=1.50s eta=19:42:46 +2026-08-25 21:10:48 | INFO | train rank=0 | step=46680/100000 progress=46.7% loss=0.047615 lr=5.956389e-06 elapsed=17:15:18 avg_step=1.33s recent_step=1.40s eta=19:42:34 +2026-08-25 21:11:00 | INFO | train rank=0 | step=46690/100000 progress=46.7% loss=0.049179 lr=5.954766e-06 elapsed=17:15:30 avg_step=1.33s recent_step=1.19s eta=19:42:19 +2026-08-25 21:11:13 | INFO | train rank=0 | step=46700/100000 progress=46.7% loss=0.052048 lr=5.953143e-06 elapsed=17:15:43 avg_step=1.33s recent_step=1.33s eta=19:42:06 +2026-08-25 21:11:25 | INFO | train rank=0 | step=46710/100000 progress=46.7% loss=0.039436 lr=5.951520e-06 elapsed=17:15:55 avg_step=1.33s recent_step=1.17s eta=19:41:50 +2026-08-25 21:11:42 | INFO | train rank=0 | step=46720/100000 progress=46.7% loss=0.057190 lr=5.949897e-06 elapsed=17:16:12 avg_step=1.33s recent_step=1.68s eta=19:41:41 +2026-08-25 21:11:53 | INFO | train rank=0 | step=46730/100000 progress=46.7% loss=0.049377 lr=5.948273e-06 elapsed=17:16:23 avg_step=1.33s recent_step=1.13s eta=19:41:26 +2026-08-25 21:12:11 | INFO | train rank=0 | step=46740/100000 progress=46.7% loss=0.058929 lr=5.946650e-06 elapsed=17:16:41 avg_step=1.33s recent_step=1.77s eta=19:41:17 +2026-08-25 21:12:22 | INFO | train rank=0 | step=46750/100000 progress=46.8% loss=0.040558 lr=5.945026e-06 elapsed=17:16:52 avg_step=1.33s recent_step=1.15s eta=19:41:02 +2026-08-25 21:12:37 | INFO | train rank=0 | step=46760/100000 progress=46.8% loss=0.042824 lr=5.943402e-06 elapsed=17:17:07 avg_step=1.33s recent_step=1.53s eta=19:40:51 +2026-08-25 21:12:51 | INFO | train rank=0 | step=46770/100000 progress=46.8% loss=0.072510 lr=5.941779e-06 elapsed=17:17:21 avg_step=1.33s recent_step=1.36s eta=19:40:38 +2026-08-25 21:13:07 | INFO | train rank=0 | step=46780/100000 progress=46.8% loss=0.048447 lr=5.940155e-06 elapsed=17:17:37 avg_step=1.33s recent_step=1.58s eta=19:40:27 +2026-08-25 21:13:22 | INFO | train rank=0 | step=46790/100000 progress=46.8% loss=0.106384 lr=5.938531e-06 elapsed=17:17:52 avg_step=1.33s recent_step=1.52s eta=19:40:16 +2026-08-25 21:13:34 | INFO | train rank=0 | step=46800/100000 progress=46.8% loss=0.057480 lr=5.936907e-06 elapsed=17:18:04 avg_step=1.33s recent_step=1.21s eta=19:40:01 +2026-08-25 21:13:46 | INFO | train rank=0 | step=46810/100000 progress=46.8% loss=0.084808 lr=5.935282e-06 elapsed=17:18:15 avg_step=1.33s recent_step=1.15s eta=19:39:46 +2026-08-25 21:13:58 | INFO | train rank=0 | step=46820/100000 progress=46.8% loss=0.049751 lr=5.933658e-06 elapsed=17:18:27 avg_step=1.33s recent_step=1.20s eta=19:39:31 +2026-08-25 21:14:09 | INFO | train rank=0 | step=46830/100000 progress=46.8% loss=0.036942 lr=5.932034e-06 elapsed=17:18:39 avg_step=1.33s recent_step=1.12s eta=19:39:16 +2026-08-25 21:14:24 | INFO | train rank=0 | step=46840/100000 progress=46.8% loss=0.058533 lr=5.930409e-06 elapsed=17:18:54 avg_step=1.33s recent_step=1.54s eta=19:39:05 +2026-08-25 21:14:38 | INFO | train rank=0 | step=46850/100000 progress=46.9% loss=0.053360 lr=5.928784e-06 elapsed=17:19:08 avg_step=1.33s recent_step=1.39s eta=19:38:52 +2026-08-25 21:14:49 | INFO | train rank=0 | step=46860/100000 progress=46.9% loss=0.025906 lr=5.927160e-06 elapsed=17:19:19 avg_step=1.33s recent_step=1.11s eta=19:38:36 +2026-08-25 21:15:02 | INFO | train rank=0 | step=46870/100000 progress=46.9% loss=0.072205 lr=5.925535e-06 elapsed=17:19:32 avg_step=1.33s recent_step=1.26s eta=19:38:22 +2026-08-25 21:15:20 | INFO | train rank=0 | step=46880/100000 progress=46.9% loss=0.052361 lr=5.923910e-06 elapsed=17:19:50 avg_step=1.33s recent_step=1.81s eta=19:38:14 +2026-08-25 21:15:32 | INFO | train rank=0 | step=46890/100000 progress=46.9% loss=0.074707 lr=5.922285e-06 elapsed=17:20:02 avg_step=1.33s recent_step=1.18s eta=19:37:59 +2026-08-25 21:15:46 | INFO | train rank=0 | step=46900/100000 progress=46.9% loss=0.038864 lr=5.920660e-06 elapsed=17:20:16 avg_step=1.33s recent_step=1.39s eta=19:37:47 +2026-08-25 21:15:57 | INFO | train rank=0 | step=46910/100000 progress=46.9% loss=0.106445 lr=5.919034e-06 elapsed=17:20:27 avg_step=1.33s recent_step=1.14s eta=19:37:31 +2026-08-25 21:16:09 | INFO | train rank=0 | step=46920/100000 progress=46.9% loss=0.062958 lr=5.917409e-06 elapsed=17:20:39 avg_step=1.33s recent_step=1.24s eta=19:37:17 +2026-08-25 21:16:21 | INFO | train rank=0 | step=46930/100000 progress=46.9% loss=0.075684 lr=5.915784e-06 elapsed=17:20:51 avg_step=1.33s recent_step=1.17s eta=19:37:02 +2026-08-25 21:16:35 | INFO | train rank=0 | step=46940/100000 progress=46.9% loss=0.046936 lr=5.914158e-06 elapsed=17:21:05 avg_step=1.33s recent_step=1.38s eta=19:36:49 +2026-08-25 21:16:47 | INFO | train rank=0 | step=46950/100000 progress=46.9% loss=0.049026 lr=5.912532e-06 elapsed=17:21:17 avg_step=1.33s recent_step=1.18s eta=19:36:34 +2026-08-25 21:17:04 | INFO | train rank=0 | step=46960/100000 progress=47.0% loss=0.033463 lr=5.910907e-06 elapsed=17:21:33 avg_step=1.33s recent_step=1.68s eta=19:36:25 +2026-08-25 21:17:18 | INFO | train rank=0 | step=46970/100000 progress=47.0% loss=0.057236 lr=5.909281e-06 elapsed=17:21:48 avg_step=1.33s recent_step=1.47s eta=19:36:13 +2026-08-25 21:17:30 | INFO | train rank=0 | step=46980/100000 progress=47.0% loss=0.043579 lr=5.907655e-06 elapsed=17:22:00 avg_step=1.33s recent_step=1.18s eta=19:35:58 +2026-08-25 21:17:42 | INFO | train rank=0 | step=46990/100000 progress=47.0% loss=0.057617 lr=5.906029e-06 elapsed=17:22:12 avg_step=1.33s recent_step=1.17s eta=19:35:43 +2026-08-25 21:17:53 | INFO | train rank=0 | step=47000/100000 progress=47.0% loss=0.051941 lr=5.904403e-06 elapsed=17:22:23 avg_step=1.33s recent_step=1.15s eta=19:35:27 +2026-08-25 21:18:08 | INFO | train rank=0 | step=47010/100000 progress=47.0% loss=0.043701 lr=5.902776e-06 elapsed=17:22:38 avg_step=1.33s recent_step=1.47s eta=19:35:16 +2026-08-25 21:18:19 | INFO | train rank=0 | step=47020/100000 progress=47.0% loss=0.045929 lr=5.901150e-06 elapsed=17:22:49 avg_step=1.33s recent_step=1.12s eta=19:35:00 +2026-08-25 21:18:33 | INFO | train rank=0 | step=47030/100000 progress=47.0% loss=0.042572 lr=5.899524e-06 elapsed=17:23:03 avg_step=1.33s recent_step=1.41s eta=19:34:48 +2026-08-25 21:18:48 | INFO | train rank=0 | step=47040/100000 progress=47.0% loss=0.041855 lr=5.897897e-06 elapsed=17:23:18 avg_step=1.33s recent_step=1.53s eta=19:34:36 +2026-08-25 21:19:00 | INFO | train rank=0 | step=47050/100000 progress=47.0% loss=0.066101 lr=5.896270e-06 elapsed=17:23:30 avg_step=1.33s recent_step=1.16s eta=19:34:21 +2026-08-25 21:19:15 | INFO | train rank=0 | step=47060/100000 progress=47.1% loss=0.110718 lr=5.894644e-06 elapsed=17:23:45 avg_step=1.33s recent_step=1.50s eta=19:34:10 +2026-08-25 21:19:26 | INFO | train rank=0 | step=47070/100000 progress=47.1% loss=0.075317 lr=5.893017e-06 elapsed=17:23:56 avg_step=1.33s recent_step=1.14s eta=19:33:54 +2026-08-25 21:19:41 | INFO | train rank=0 | step=47080/100000 progress=47.1% loss=0.110046 lr=5.891390e-06 elapsed=17:24:11 avg_step=1.33s recent_step=1.49s eta=19:33:43 +2026-08-25 21:19:54 | INFO | train rank=0 | step=47090/100000 progress=47.1% loss=0.047089 lr=5.889763e-06 elapsed=17:24:24 avg_step=1.33s recent_step=1.26s eta=19:33:29 +2026-08-25 21:20:06 | INFO | train rank=0 | step=47100/100000 progress=47.1% loss=0.047989 lr=5.888136e-06 elapsed=17:24:36 avg_step=1.33s recent_step=1.17s eta=19:33:14 +2026-08-25 21:20:19 | INFO | train rank=0 | step=47110/100000 progress=47.1% loss=0.054901 lr=5.886509e-06 elapsed=17:24:49 avg_step=1.33s recent_step=1.31s eta=19:33:00 +2026-08-25 21:20:35 | INFO | train rank=0 | step=47120/100000 progress=47.1% loss=0.068481 lr=5.884881e-06 elapsed=17:25:05 avg_step=1.33s recent_step=1.63s eta=19:32:50 +2026-08-25 21:20:46 | INFO | train rank=0 | step=47130/100000 progress=47.1% loss=0.151001 lr=5.883254e-06 elapsed=17:25:16 avg_step=1.33s recent_step=1.12s eta=19:32:35 +2026-08-25 21:21:02 | INFO | train rank=0 | step=47140/100000 progress=47.1% loss=0.042236 lr=5.881626e-06 elapsed=17:25:32 avg_step=1.33s recent_step=1.54s eta=19:32:24 +2026-08-25 21:21:14 | INFO | train rank=0 | step=47150/100000 progress=47.1% loss=0.057556 lr=5.879999e-06 elapsed=17:25:44 avg_step=1.33s recent_step=1.23s eta=19:32:09 +2026-08-25 21:21:29 | INFO | train rank=0 | step=47160/100000 progress=47.2% loss=0.046021 lr=5.878371e-06 elapsed=17:25:59 avg_step=1.33s recent_step=1.50s eta=19:31:58 +2026-08-25 21:21:41 | INFO | train rank=0 | step=47170/100000 progress=47.2% loss=0.035385 lr=5.876743e-06 elapsed=17:26:11 avg_step=1.33s recent_step=1.23s eta=19:31:43 +2026-08-25 21:21:53 | INFO | train rank=0 | step=47180/100000 progress=47.2% loss=0.047104 lr=5.875115e-06 elapsed=17:26:23 avg_step=1.33s recent_step=1.14s eta=19:31:28 +2026-08-25 21:22:07 | INFO | train rank=0 | step=47190/100000 progress=47.2% loss=0.047180 lr=5.873487e-06 elapsed=17:26:37 avg_step=1.33s recent_step=1.43s eta=19:31:16 +2026-08-25 21:22:18 | INFO | train rank=0 | step=47200/100000 progress=47.2% loss=0.048126 lr=5.871859e-06 elapsed=17:26:48 avg_step=1.33s recent_step=1.12s eta=19:31:00 +2026-08-25 21:22:30 | INFO | train rank=0 | step=47210/100000 progress=47.2% loss=0.050598 lr=5.870231e-06 elapsed=17:27:00 avg_step=1.33s recent_step=1.17s eta=19:30:45 +2026-08-25 21:22:48 | INFO | train rank=0 | step=47220/100000 progress=47.2% loss=0.048203 lr=5.868603e-06 elapsed=17:27:18 avg_step=1.33s recent_step=1.83s eta=19:30:37 +2026-08-25 21:23:03 | INFO | train rank=0 | step=47230/100000 progress=47.2% loss=0.047699 lr=5.866974e-06 elapsed=17:27:33 avg_step=1.33s recent_step=1.45s eta=19:30:25 +2026-08-25 21:23:16 | INFO | train rank=0 | step=47240/100000 progress=47.2% loss=0.048401 lr=5.865346e-06 elapsed=17:27:46 avg_step=1.33s recent_step=1.34s eta=19:30:12 +2026-08-25 21:23:29 | INFO | train rank=0 | step=47250/100000 progress=47.2% loss=0.039719 lr=5.863717e-06 elapsed=17:27:59 avg_step=1.33s recent_step=1.28s eta=19:29:58 +2026-08-25 21:23:46 | INFO | train rank=0 | step=47260/100000 progress=47.3% loss=0.059235 lr=5.862089e-06 elapsed=17:28:15 avg_step=1.33s recent_step=1.65s eta=19:29:48 +2026-08-25 21:23:57 | INFO | train rank=0 | step=47270/100000 progress=47.3% loss=0.059296 lr=5.860460e-06 elapsed=17:28:27 avg_step=1.33s recent_step=1.19s eta=19:29:34 +2026-08-25 21:24:09 | INFO | train rank=0 | step=47280/100000 progress=47.3% loss=0.103455 lr=5.858831e-06 elapsed=17:28:39 avg_step=1.33s recent_step=1.15s eta=19:29:18 +2026-08-25 21:24:21 | INFO | train rank=0 | step=47290/100000 progress=47.3% loss=0.038239 lr=5.857202e-06 elapsed=17:28:51 avg_step=1.33s recent_step=1.18s eta=19:29:03 +2026-08-25 21:24:32 | INFO | train rank=0 | step=47300/100000 progress=47.3% loss=0.066132 lr=5.855573e-06 elapsed=17:29:02 avg_step=1.33s recent_step=1.17s eta=19:28:48 +2026-08-25 21:24:50 | INFO | train rank=0 | step=47310/100000 progress=47.3% loss=0.029690 lr=5.853944e-06 elapsed=17:29:19 avg_step=1.33s recent_step=1.71s eta=19:28:39 +2026-08-25 21:25:02 | INFO | train rank=0 | step=47320/100000 progress=47.3% loss=0.066803 lr=5.852315e-06 elapsed=17:29:32 avg_step=1.33s recent_step=1.20s eta=19:28:24 +2026-08-25 21:25:14 | INFO | train rank=0 | step=47330/100000 progress=47.3% loss=0.072739 lr=5.850685e-06 elapsed=17:29:44 avg_step=1.33s recent_step=1.22s eta=19:28:10 +2026-08-25 21:25:29 | INFO | train rank=0 | step=47340/100000 progress=47.3% loss=0.052795 lr=5.849056e-06 elapsed=17:29:59 avg_step=1.33s recent_step=1.50s eta=19:27:58 +2026-08-25 21:25:43 | INFO | train rank=0 | step=47350/100000 progress=47.3% loss=0.065079 lr=5.847427e-06 elapsed=17:30:13 avg_step=1.33s recent_step=1.43s eta=19:27:46 +2026-08-25 21:25:54 | INFO | train rank=0 | step=47360/100000 progress=47.4% loss=0.070007 lr=5.845797e-06 elapsed=17:30:24 avg_step=1.33s recent_step=1.12s eta=19:27:31 +2026-08-25 21:26:09 | INFO | train rank=0 | step=47370/100000 progress=47.4% loss=0.059982 lr=5.844167e-06 elapsed=17:30:39 avg_step=1.33s recent_step=1.45s eta=19:27:19 +2026-08-25 21:26:21 | INFO | train rank=0 | step=47380/100000 progress=47.4% loss=0.052185 lr=5.842537e-06 elapsed=17:30:51 avg_step=1.33s recent_step=1.26s eta=19:27:05 +2026-08-25 21:26:35 | INFO | train rank=0 | step=47390/100000 progress=47.4% loss=0.061310 lr=5.840908e-06 elapsed=17:31:05 avg_step=1.33s recent_step=1.36s eta=19:26:52 +2026-08-25 21:26:49 | INFO | train rank=0 | step=47400/100000 progress=47.4% loss=0.036659 lr=5.839278e-06 elapsed=17:31:19 avg_step=1.33s recent_step=1.36s eta=19:26:39 +2026-08-25 21:27:00 | INFO | train rank=0 | step=47410/100000 progress=47.4% loss=0.060211 lr=5.837648e-06 elapsed=17:31:30 avg_step=1.33s recent_step=1.17s eta=19:26:24 +2026-08-25 21:27:17 | INFO | train rank=0 | step=47420/100000 progress=47.4% loss=0.031364 lr=5.836017e-06 elapsed=17:31:47 avg_step=1.33s recent_step=1.65s eta=19:26:14 +2026-08-25 21:27:28 | INFO | train rank=0 | step=47430/100000 progress=47.4% loss=0.056061 lr=5.834387e-06 elapsed=17:31:58 avg_step=1.33s recent_step=1.16s eta=19:25:59 +2026-08-25 21:27:44 | INFO | train rank=0 | step=47440/100000 progress=47.4% loss=0.053741 lr=5.832757e-06 elapsed=17:32:14 avg_step=1.33s recent_step=1.55s eta=19:25:48 +2026-08-25 21:27:56 | INFO | train rank=0 | step=47450/100000 progress=47.4% loss=0.068878 lr=5.831126e-06 elapsed=17:32:26 avg_step=1.33s recent_step=1.19s eta=19:25:33 +2026-08-25 21:28:09 | INFO | train rank=0 | step=47460/100000 progress=47.5% loss=0.042328 lr=5.829496e-06 elapsed=17:32:39 avg_step=1.33s recent_step=1.32s eta=19:25:19 +2026-08-25 21:28:21 | INFO | train rank=0 | step=47470/100000 progress=47.5% loss=0.039017 lr=5.827865e-06 elapsed=17:32:51 avg_step=1.33s recent_step=1.22s eta=19:25:05 +2026-08-25 21:28:33 | INFO | train rank=0 | step=47480/100000 progress=47.5% loss=0.052551 lr=5.826235e-06 elapsed=17:33:03 avg_step=1.33s recent_step=1.17s eta=19:24:50 +2026-08-25 21:28:48 | INFO | train rank=0 | step=47490/100000 progress=47.5% loss=0.065948 lr=5.824604e-06 elapsed=17:33:17 avg_step=1.33s recent_step=1.47s eta=19:24:38 +2026-08-25 21:28:59 | INFO | train rank=0 | step=47500/100000 progress=47.5% loss=0.047134 lr=5.822973e-06 elapsed=17:33:29 avg_step=1.33s recent_step=1.16s eta=19:24:23 +2026-08-25 21:29:15 | INFO | train rank=0 | step=47510/100000 progress=47.5% loss=0.037140 lr=5.821342e-06 elapsed=17:33:45 avg_step=1.33s recent_step=1.60s eta=19:24:12 +2026-08-25 21:29:27 | INFO | train rank=0 | step=47520/100000 progress=47.5% loss=0.049774 lr=5.819711e-06 elapsed=17:33:57 avg_step=1.33s recent_step=1.15s eta=19:23:57 +2026-08-25 21:29:40 | INFO | train rank=0 | step=47530/100000 progress=47.5% loss=0.062813 lr=5.818080e-06 elapsed=17:34:10 avg_step=1.33s recent_step=1.35s eta=19:23:44 +2026-08-25 21:29:54 | INFO | train rank=0 | step=47540/100000 progress=47.5% loss=0.071808 lr=5.816449e-06 elapsed=17:34:24 avg_step=1.33s recent_step=1.42s eta=19:23:32 +2026-08-25 21:30:11 | INFO | train rank=0 | step=47550/100000 progress=47.5% loss=0.058136 lr=5.814817e-06 elapsed=17:34:41 avg_step=1.33s recent_step=1.62s eta=19:23:22 +2026-08-25 21:30:22 | INFO | train rank=0 | step=47560/100000 progress=47.6% loss=0.046722 lr=5.813186e-06 elapsed=17:34:52 avg_step=1.33s recent_step=1.16s eta=19:23:06 +2026-08-25 21:30:37 | INFO | train rank=0 | step=47570/100000 progress=47.6% loss=0.071991 lr=5.811554e-06 elapsed=17:35:07 avg_step=1.33s recent_step=1.52s eta=19:22:55 +2026-08-25 21:30:49 | INFO | train rank=0 | step=47580/100000 progress=47.6% loss=0.072510 lr=5.809923e-06 elapsed=17:35:19 avg_step=1.33s recent_step=1.15s eta=19:22:40 +2026-08-25 21:31:02 | INFO | train rank=0 | step=47590/100000 progress=47.6% loss=0.043945 lr=5.808291e-06 elapsed=17:35:32 avg_step=1.33s recent_step=1.34s eta=19:22:27 +2026-08-25 21:31:16 | INFO | train rank=0 | step=47600/100000 progress=47.6% loss=0.050934 lr=5.806659e-06 elapsed=17:35:46 avg_step=1.33s recent_step=1.37s eta=19:22:14 +2026-08-25 21:31:27 | INFO | train rank=0 | step=47610/100000 progress=47.6% loss=0.044968 lr=5.805027e-06 elapsed=17:35:57 avg_step=1.33s recent_step=1.14s eta=19:21:58 +2026-08-25 21:31:43 | INFO | train rank=0 | step=47620/100000 progress=47.6% loss=0.048676 lr=5.803395e-06 elapsed=17:36:13 avg_step=1.33s recent_step=1.58s eta=19:21:48 +2026-08-25 21:31:59 | INFO | train rank=0 | step=47630/100000 progress=47.6% loss=0.047791 lr=5.801763e-06 elapsed=17:36:29 avg_step=1.33s recent_step=1.55s eta=19:21:37 +2026-08-25 21:32:11 | INFO | train rank=0 | step=47640/100000 progress=47.6% loss=0.039474 lr=5.800131e-06 elapsed=17:36:41 avg_step=1.33s recent_step=1.23s eta=19:21:23 +2026-08-25 21:32:30 | INFO | train rank=0 | step=47650/100000 progress=47.6% loss=0.047104 lr=5.798499e-06 elapsed=17:37:00 avg_step=1.33s recent_step=1.90s eta=19:21:15 +2026-08-25 21:32:42 | INFO | train rank=0 | step=47660/100000 progress=47.7% loss=0.073517 lr=5.796867e-06 elapsed=17:37:12 avg_step=1.33s recent_step=1.17s eta=19:21:00 +2026-08-25 21:32:54 | INFO | train rank=0 | step=47670/100000 progress=47.7% loss=0.047470 lr=5.795234e-06 elapsed=17:37:23 avg_step=1.33s recent_step=1.18s eta=19:20:45 +2026-08-25 21:33:05 | INFO | train rank=0 | step=47680/100000 progress=47.7% loss=0.042603 lr=5.793602e-06 elapsed=17:37:35 avg_step=1.33s recent_step=1.15s eta=19:20:30 +2026-08-25 21:33:18 | INFO | train rank=0 | step=47690/100000 progress=47.7% loss=0.054672 lr=5.791969e-06 elapsed=17:37:48 avg_step=1.33s recent_step=1.29s eta=19:20:16 +2026-08-25 21:33:29 | INFO | train rank=0 | step=47700/100000 progress=47.7% loss=0.062592 lr=5.790337e-06 elapsed=17:37:59 avg_step=1.33s recent_step=1.15s eta=19:20:01 +2026-08-25 21:33:48 | INFO | train rank=0 | step=47710/100000 progress=47.7% loss=0.041336 lr=5.788704e-06 elapsed=17:38:18 avg_step=1.33s recent_step=1.88s eta=19:19:54 +2026-08-25 21:34:00 | INFO | train rank=0 | step=47720/100000 progress=47.7% loss=0.039673 lr=5.787071e-06 elapsed=17:38:30 avg_step=1.33s recent_step=1.19s eta=19:19:39 +2026-08-25 21:34:12 | INFO | train rank=0 | step=47730/100000 progress=47.7% loss=0.046066 lr=5.785438e-06 elapsed=17:38:42 avg_step=1.33s recent_step=1.18s eta=19:19:24 +2026-08-25 21:34:23 | INFO | train rank=0 | step=47740/100000 progress=47.7% loss=0.086456 lr=5.783805e-06 elapsed=17:38:53 avg_step=1.33s recent_step=1.13s eta=19:19:08 +2026-08-25 21:34:41 | INFO | train rank=0 | step=47750/100000 progress=47.8% loss=0.041565 lr=5.782172e-06 elapsed=17:39:11 avg_step=1.33s recent_step=1.83s eta=19:19:01 +2026-08-25 21:34:53 | INFO | train rank=0 | step=47760/100000 progress=47.8% loss=0.059875 lr=5.780539e-06 elapsed=17:39:23 avg_step=1.33s recent_step=1.12s eta=19:18:45 +2026-08-25 21:35:06 | INFO | train rank=0 | step=47770/100000 progress=47.8% loss=0.046768 lr=5.778906e-06 elapsed=17:39:36 avg_step=1.33s recent_step=1.35s eta=19:18:32 +2026-08-25 21:35:18 | INFO | train rank=0 | step=47780/100000 progress=47.8% loss=0.044861 lr=5.777273e-06 elapsed=17:39:48 avg_step=1.33s recent_step=1.16s eta=19:18:17 +2026-08-25 21:35:29 | INFO | train rank=0 | step=47790/100000 progress=47.8% loss=0.055573 lr=5.775639e-06 elapsed=17:39:59 avg_step=1.33s recent_step=1.15s eta=19:18:01 +2026-08-25 21:35:48 | INFO | train rank=0 | step=47800/100000 progress=47.8% loss=0.054810 lr=5.774006e-06 elapsed=17:40:18 avg_step=1.33s recent_step=1.85s eta=19:17:54 +2026-08-25 21:35:59 | INFO | train rank=0 | step=47810/100000 progress=47.8% loss=0.051163 lr=5.772372e-06 elapsed=17:40:29 avg_step=1.33s recent_step=1.12s eta=19:17:38 +2026-08-25 21:36:11 | INFO | train rank=0 | step=47820/100000 progress=47.8% loss=0.037941 lr=5.770738e-06 elapsed=17:40:41 avg_step=1.33s recent_step=1.18s eta=19:17:23 +2026-08-25 21:36:27 | INFO | train rank=0 | step=47830/100000 progress=47.8% loss=0.060303 lr=5.769105e-06 elapsed=17:40:57 avg_step=1.33s recent_step=1.63s eta=19:17:13 +2026-08-25 21:36:39 | INFO | train rank=0 | step=47840/100000 progress=47.8% loss=0.038193 lr=5.767471e-06 elapsed=17:41:09 avg_step=1.33s recent_step=1.19s eta=19:16:58 +2026-08-25 21:36:56 | INFO | train rank=0 | step=47850/100000 progress=47.9% loss=0.049080 lr=5.765837e-06 elapsed=17:41:26 avg_step=1.33s recent_step=1.67s eta=19:16:49 +2026-08-25 21:37:07 | INFO | train rank=0 | step=47860/100000 progress=47.9% loss=0.049774 lr=5.764203e-06 elapsed=17:41:37 avg_step=1.33s recent_step=1.15s eta=19:16:33 +2026-08-25 21:37:21 | INFO | train rank=0 | step=47870/100000 progress=47.9% loss=0.076416 lr=5.762569e-06 elapsed=17:41:51 avg_step=1.33s recent_step=1.43s eta=19:16:21 +2026-08-25 21:37:33 | INFO | train rank=0 | step=47880/100000 progress=47.9% loss=0.059204 lr=5.760935e-06 elapsed=17:42:03 avg_step=1.33s recent_step=1.16s eta=19:16:06 +2026-08-25 21:37:45 | INFO | train rank=0 | step=47890/100000 progress=47.9% loss=0.088440 lr=5.759300e-06 elapsed=17:42:15 avg_step=1.33s recent_step=1.22s eta=19:15:51 +2026-08-25 21:38:01 | INFO | train rank=0 | step=47900/100000 progress=47.9% loss=0.037140 lr=5.757666e-06 elapsed=17:42:31 avg_step=1.33s recent_step=1.55s eta=19:15:41 +2026-08-25 21:38:16 | INFO | train rank=0 | step=47910/100000 progress=47.9% loss=0.097961 lr=5.756032e-06 elapsed=17:42:46 avg_step=1.33s recent_step=1.54s eta=19:15:30 +2026-08-25 21:38:28 | INFO | train rank=0 | step=47920/100000 progress=47.9% loss=0.053848 lr=5.754397e-06 elapsed=17:42:58 avg_step=1.33s recent_step=1.16s eta=19:15:14 +2026-08-25 21:38:39 | INFO | train rank=0 | step=47930/100000 progress=47.9% loss=0.067078 lr=5.752763e-06 elapsed=17:43:09 avg_step=1.33s recent_step=1.14s eta=19:14:59 +2026-08-25 21:38:52 | INFO | train rank=0 | step=47940/100000 progress=47.9% loss=0.086426 lr=5.751128e-06 elapsed=17:43:22 avg_step=1.33s recent_step=1.26s eta=19:14:45 +2026-08-25 21:39:03 | INFO | train rank=0 | step=47950/100000 progress=47.9% loss=0.036621 lr=5.749493e-06 elapsed=17:43:33 avg_step=1.33s recent_step=1.17s eta=19:14:30 +2026-08-25 21:39:17 | INFO | train rank=0 | step=47960/100000 progress=48.0% loss=0.039886 lr=5.747858e-06 elapsed=17:43:47 avg_step=1.33s recent_step=1.32s eta=19:14:16 +2026-08-25 21:39:31 | INFO | train rank=0 | step=47970/100000 progress=48.0% loss=0.046860 lr=5.746223e-06 elapsed=17:44:01 avg_step=1.33s recent_step=1.48s eta=19:14:05 +2026-08-25 21:39:46 | INFO | train rank=0 | step=47980/100000 progress=48.0% loss=0.035080 lr=5.744588e-06 elapsed=17:44:16 avg_step=1.33s recent_step=1.47s eta=19:13:53 +2026-08-25 21:40:01 | INFO | train rank=0 | step=47990/100000 progress=48.0% loss=0.068481 lr=5.742953e-06 elapsed=17:44:31 avg_step=1.33s recent_step=1.50s eta=19:13:41 +2026-08-25 21:40:12 | INFO | train rank=0 | step=48000/100000 progress=48.0% loss=0.055603 lr=5.741318e-06 elapsed=17:44:42 avg_step=1.33s recent_step=1.14s eta=19:13:26 +2026-08-25 21:40:27 | INFO | train rank=0 | step=48010/100000 progress=48.0% loss=0.051514 lr=5.739683e-06 elapsed=17:44:57 avg_step=1.33s recent_step=1.42s eta=19:13:14 +2026-08-25 21:40:40 | INFO | train rank=0 | step=48020/100000 progress=48.0% loss=0.058014 lr=5.738048e-06 elapsed=17:45:09 avg_step=1.33s recent_step=1.28s eta=19:13:00 +2026-08-25 21:40:54 | INFO | train rank=0 | step=48030/100000 progress=48.0% loss=0.046509 lr=5.736412e-06 elapsed=17:45:24 avg_step=1.33s recent_step=1.47s eta=19:12:48 +2026-08-25 21:41:10 | INFO | train rank=0 | step=48040/100000 progress=48.0% loss=0.049240 lr=5.734777e-06 elapsed=17:45:39 avg_step=1.33s recent_step=1.54s eta=19:12:37 +2026-08-25 21:41:21 | INFO | train rank=0 | step=48050/100000 progress=48.0% loss=0.038651 lr=5.733141e-06 elapsed=17:45:51 avg_step=1.33s recent_step=1.12s eta=19:12:21 +2026-08-25 21:41:36 | INFO | train rank=0 | step=48060/100000 progress=48.1% loss=0.047974 lr=5.731506e-06 elapsed=17:46:06 avg_step=1.33s recent_step=1.51s eta=19:12:10 +2026-08-25 21:41:50 | INFO | train rank=0 | step=48070/100000 progress=48.1% loss=0.056564 lr=5.729870e-06 elapsed=17:46:20 avg_step=1.33s recent_step=1.44s eta=19:11:58 +2026-08-25 21:42:05 | INFO | train rank=0 | step=48080/100000 progress=48.1% loss=0.082703 lr=5.728234e-06 elapsed=17:46:35 avg_step=1.33s recent_step=1.45s eta=19:11:46 +2026-08-25 21:42:16 | INFO | train rank=0 | step=48090/100000 progress=48.1% loss=0.064484 lr=5.726598e-06 elapsed=17:46:46 avg_step=1.33s recent_step=1.15s eta=19:11:31 +2026-08-25 21:42:28 | INFO | train rank=0 | step=48100/100000 progress=48.1% loss=0.080109 lr=5.724962e-06 elapsed=17:46:58 avg_step=1.33s recent_step=1.13s eta=19:11:15 +2026-08-25 21:42:40 | INFO | train rank=0 | step=48110/100000 progress=48.1% loss=0.051880 lr=5.723326e-06 elapsed=17:47:10 avg_step=1.33s recent_step=1.24s eta=19:11:01 +2026-08-25 21:42:52 | INFO | train rank=0 | step=48120/100000 progress=48.1% loss=0.047379 lr=5.721690e-06 elapsed=17:47:22 avg_step=1.33s recent_step=1.21s eta=19:10:46 +2026-08-25 21:43:06 | INFO | train rank=0 | step=48130/100000 progress=48.1% loss=0.061066 lr=5.720054e-06 elapsed=17:47:36 avg_step=1.33s recent_step=1.41s eta=19:10:34 +2026-08-25 21:43:18 | INFO | train rank=0 | step=48140/100000 progress=48.1% loss=0.034584 lr=5.718418e-06 elapsed=17:47:48 avg_step=1.33s recent_step=1.16s eta=19:10:19 +2026-08-25 21:43:30 | INFO | train rank=0 | step=48150/100000 progress=48.1% loss=0.048996 lr=5.716781e-06 elapsed=17:48:00 avg_step=1.33s recent_step=1.25s eta=19:10:04 +2026-08-25 21:43:46 | INFO | train rank=0 | step=48160/100000 progress=48.2% loss=0.037781 lr=5.715145e-06 elapsed=17:48:15 avg_step=1.33s recent_step=1.53s eta=19:09:53 +2026-08-25 21:44:01 | INFO | train rank=0 | step=48170/100000 progress=48.2% loss=0.041702 lr=5.713508e-06 elapsed=17:48:31 avg_step=1.33s recent_step=1.55s eta=19:09:42 +2026-08-25 21:44:12 | INFO | train rank=0 | step=48180/100000 progress=48.2% loss=0.032776 lr=5.711872e-06 elapsed=17:48:42 avg_step=1.33s recent_step=1.12s eta=19:09:27 +2026-08-25 21:44:24 | INFO | train rank=0 | step=48190/100000 progress=48.2% loss=0.045349 lr=5.710235e-06 elapsed=17:48:54 avg_step=1.33s recent_step=1.15s eta=19:09:12 +2026-08-25 21:44:38 | INFO | train rank=0 | step=48200/100000 progress=48.2% loss=0.036224 lr=5.708598e-06 elapsed=17:49:08 avg_step=1.33s recent_step=1.45s eta=19:08:59 +2026-08-25 21:44:50 | INFO | train rank=0 | step=48210/100000 progress=48.2% loss=0.055771 lr=5.706962e-06 elapsed=17:49:20 avg_step=1.33s recent_step=1.13s eta=19:08:44 +2026-08-25 21:45:07 | INFO | train rank=0 | step=48220/100000 progress=48.2% loss=0.067230 lr=5.705325e-06 elapsed=17:49:37 avg_step=1.33s recent_step=1.75s eta=19:08:35 +2026-08-25 21:45:19 | INFO | train rank=0 | step=48230/100000 progress=48.2% loss=0.044914 lr=5.703688e-06 elapsed=17:49:49 avg_step=1.33s recent_step=1.17s eta=19:08:20 +2026-08-25 21:45:37 | INFO | train rank=0 | step=48240/100000 progress=48.2% loss=0.099060 lr=5.702051e-06 elapsed=17:50:07 avg_step=1.33s recent_step=1.83s eta=19:08:12 +2026-08-25 21:45:51 | INFO | train rank=0 | step=48250/100000 progress=48.2% loss=0.053070 lr=5.700413e-06 elapsed=17:50:20 avg_step=1.33s recent_step=1.35s eta=19:07:59 +2026-08-25 21:46:02 | INFO | train rank=0 | step=48260/100000 progress=48.3% loss=0.090393 lr=5.698776e-06 elapsed=17:50:32 avg_step=1.33s recent_step=1.14s eta=19:07:44 +2026-08-25 21:46:17 | INFO | train rank=0 | step=48270/100000 progress=48.3% loss=0.051605 lr=5.697139e-06 elapsed=17:50:47 avg_step=1.33s recent_step=1.48s eta=19:07:32 +2026-08-25 21:46:31 | INFO | train rank=0 | step=48280/100000 progress=48.3% loss=0.045593 lr=5.695502e-06 elapsed=17:51:01 avg_step=1.33s recent_step=1.39s eta=19:07:19 +2026-08-25 21:46:45 | INFO | train rank=0 | step=48290/100000 progress=48.3% loss=0.089325 lr=5.693864e-06 elapsed=17:51:15 avg_step=1.33s recent_step=1.44s eta=19:07:07 +2026-08-25 21:46:57 | INFO | train rank=0 | step=48300/100000 progress=48.3% loss=0.106018 lr=5.692227e-06 elapsed=17:51:27 avg_step=1.33s recent_step=1.21s eta=19:06:52 +2026-08-25 21:47:11 | INFO | train rank=0 | step=48310/100000 progress=48.3% loss=0.078857 lr=5.690589e-06 elapsed=17:51:41 avg_step=1.33s recent_step=1.42s eta=19:06:40 +2026-08-25 21:47:25 | INFO | train rank=0 | step=48320/100000 progress=48.3% loss=0.061127 lr=5.688951e-06 elapsed=17:51:55 avg_step=1.33s recent_step=1.37s eta=19:06:27 +2026-08-25 21:47:41 | INFO | train rank=0 | step=48330/100000 progress=48.3% loss=0.047485 lr=5.687314e-06 elapsed=17:52:11 avg_step=1.33s recent_step=1.59s eta=19:06:17 +2026-08-25 21:47:52 | INFO | train rank=0 | step=48340/100000 progress=48.3% loss=0.053314 lr=5.685676e-06 elapsed=17:52:22 avg_step=1.33s recent_step=1.09s eta=19:06:01 +2026-08-25 21:48:07 | INFO | train rank=0 | step=48350/100000 progress=48.4% loss=0.046532 lr=5.684038e-06 elapsed=17:52:37 avg_step=1.33s recent_step=1.52s eta=19:05:49 +2026-08-25 21:48:18 | INFO | train rank=0 | step=48360/100000 progress=48.4% loss=0.042465 lr=5.682400e-06 elapsed=17:52:48 avg_step=1.33s recent_step=1.12s eta=19:05:34 +2026-08-25 21:48:30 | INFO | train rank=0 | step=48370/100000 progress=48.4% loss=0.070587 lr=5.680762e-06 elapsed=17:53:00 avg_step=1.33s recent_step=1.18s eta=19:05:19 +2026-08-25 21:48:42 | INFO | train rank=0 | step=48380/100000 progress=48.4% loss=0.057449 lr=5.679124e-06 elapsed=17:53:12 avg_step=1.33s recent_step=1.21s eta=19:05:04 +2026-08-25 21:48:54 | INFO | train rank=0 | step=48390/100000 progress=48.4% loss=0.031158 lr=5.677486e-06 elapsed=17:53:24 avg_step=1.33s recent_step=1.19s eta=19:04:49 +2026-08-25 21:49:10 | INFO | train rank=0 | step=48400/100000 progress=48.4% loss=0.060974 lr=5.675847e-06 elapsed=17:53:40 avg_step=1.33s recent_step=1.63s eta=19:04:39 +2026-08-25 21:49:25 | INFO | train rank=0 | step=48410/100000 progress=48.4% loss=0.071533 lr=5.674209e-06 elapsed=17:53:55 avg_step=1.33s recent_step=1.47s eta=19:04:27 +2026-08-25 21:49:36 | INFO | train rank=0 | step=48420/100000 progress=48.4% loss=0.052155 lr=5.672571e-06 elapsed=17:54:06 avg_step=1.33s recent_step=1.14s eta=19:04:12 +2026-08-25 21:49:50 | INFO | train rank=0 | step=48430/100000 progress=48.4% loss=0.075836 lr=5.670932e-06 elapsed=17:54:20 avg_step=1.33s recent_step=1.38s eta=19:03:59 +2026-08-25 21:50:03 | INFO | train rank=0 | step=48440/100000 progress=48.4% loss=0.054016 lr=5.669294e-06 elapsed=17:54:33 avg_step=1.33s recent_step=1.31s eta=19:03:46 +2026-08-25 21:50:15 | INFO | train rank=0 | step=48450/100000 progress=48.4% loss=0.060791 lr=5.667655e-06 elapsed=17:54:45 avg_step=1.33s recent_step=1.19s eta=19:03:31 +2026-08-25 21:50:30 | INFO | train rank=0 | step=48460/100000 progress=48.5% loss=0.055084 lr=5.666016e-06 elapsed=17:55:00 avg_step=1.33s recent_step=1.48s eta=19:03:19 +2026-08-25 21:50:42 | INFO | train rank=0 | step=48470/100000 progress=48.5% loss=0.037704 lr=5.664378e-06 elapsed=17:55:12 avg_step=1.33s recent_step=1.20s eta=19:03:04 +2026-08-25 21:50:53 | INFO | train rank=0 | step=48480/100000 progress=48.5% loss=0.056122 lr=5.662739e-06 elapsed=17:55:23 avg_step=1.33s recent_step=1.16s eta=19:02:49 +2026-08-25 21:51:07 | INFO | train rank=0 | step=48490/100000 progress=48.5% loss=0.045685 lr=5.661100e-06 elapsed=17:55:37 avg_step=1.33s recent_step=1.39s eta=19:02:37 +2026-08-25 21:51:19 | INFO | train rank=0 | step=48500/100000 progress=48.5% loss=0.061096 lr=5.659461e-06 elapsed=17:55:48 avg_step=1.33s recent_step=1.13s eta=19:02:21 +2026-08-25 21:51:30 | INFO | train rank=0 | step=48510/100000 progress=48.5% loss=0.059326 lr=5.657822e-06 elapsed=17:56:00 avg_step=1.33s recent_step=1.17s eta=19:02:06 +2026-08-25 21:51:42 | INFO | train rank=0 | step=48520/100000 progress=48.5% loss=0.064453 lr=5.656183e-06 elapsed=17:56:12 avg_step=1.33s recent_step=1.16s eta=19:01:51 +2026-08-25 21:51:54 | INFO | train rank=0 | step=48530/100000 progress=48.5% loss=0.044830 lr=5.654543e-06 elapsed=17:56:24 avg_step=1.33s recent_step=1.17s eta=19:01:36 +2026-08-25 21:52:09 | INFO | train rank=0 | step=48540/100000 progress=48.5% loss=0.058975 lr=5.652904e-06 elapsed=17:56:39 avg_step=1.33s recent_step=1.55s eta=19:01:25 +2026-08-25 21:52:21 | INFO | train rank=0 | step=48550/100000 progress=48.5% loss=0.034416 lr=5.651265e-06 elapsed=17:56:51 avg_step=1.33s recent_step=1.17s eta=19:01:10 +2026-08-25 21:52:35 | INFO | train rank=0 | step=48560/100000 progress=48.6% loss=0.082153 lr=5.649625e-06 elapsed=17:57:05 avg_step=1.33s recent_step=1.41s eta=19:00:58 +2026-08-25 21:52:48 | INFO | train rank=0 | step=48570/100000 progress=48.6% loss=0.066589 lr=5.647986e-06 elapsed=17:57:18 avg_step=1.33s recent_step=1.28s eta=19:00:44 +2026-08-25 21:53:02 | INFO | train rank=0 | step=48580/100000 progress=48.6% loss=0.099304 lr=5.646346e-06 elapsed=17:57:32 avg_step=1.33s recent_step=1.45s eta=19:00:32 +2026-08-25 21:53:14 | INFO | train rank=0 | step=48590/100000 progress=48.6% loss=0.068283 lr=5.644707e-06 elapsed=17:57:44 avg_step=1.33s recent_step=1.14s eta=19:00:16 +2026-08-25 21:53:29 | INFO | train rank=0 | step=48600/100000 progress=48.6% loss=0.066254 lr=5.643067e-06 elapsed=17:57:59 avg_step=1.33s recent_step=1.52s eta=19:00:05 +2026-08-25 21:53:41 | INFO | train rank=0 | step=48610/100000 progress=48.6% loss=0.037598 lr=5.641427e-06 elapsed=17:58:11 avg_step=1.33s recent_step=1.22s eta=18:59:51 +2026-08-25 21:53:53 | INFO | train rank=0 | step=48620/100000 progress=48.6% loss=0.054916 lr=5.639787e-06 elapsed=17:58:22 avg_step=1.33s recent_step=1.16s eta=18:59:35 +2026-08-25 21:54:05 | INFO | train rank=0 | step=48630/100000 progress=48.6% loss=0.051682 lr=5.638148e-06 elapsed=17:58:35 avg_step=1.33s recent_step=1.26s eta=18:59:21 +2026-08-25 21:54:21 | INFO | train rank=0 | step=48640/100000 progress=48.6% loss=0.045166 lr=5.636508e-06 elapsed=17:58:51 avg_step=1.33s recent_step=1.56s eta=18:59:10 +2026-08-25 21:54:36 | INFO | train rank=0 | step=48650/100000 progress=48.6% loss=0.062958 lr=5.634867e-06 elapsed=17:59:06 avg_step=1.33s recent_step=1.55s eta=18:58:59 +2026-08-25 21:54:51 | INFO | train rank=0 | step=48660/100000 progress=48.7% loss=0.073547 lr=5.633227e-06 elapsed=17:59:20 avg_step=1.33s recent_step=1.44s eta=18:58:47 +2026-08-25 21:55:03 | INFO | train rank=0 | step=48670/100000 progress=48.7% loss=0.069794 lr=5.631587e-06 elapsed=17:59:33 avg_step=1.33s recent_step=1.22s eta=18:58:33 +2026-08-25 21:55:17 | INFO | train rank=0 | step=48680/100000 progress=48.7% loss=0.070953 lr=5.629947e-06 elapsed=17:59:47 avg_step=1.33s recent_step=1.44s eta=18:58:21 +2026-08-25 21:55:33 | INFO | train rank=0 | step=48690/100000 progress=48.7% loss=0.137878 lr=5.628307e-06 elapsed=18:00:03 avg_step=1.33s recent_step=1.63s eta=18:58:10 +2026-08-25 21:55:45 | INFO | train rank=0 | step=48700/100000 progress=48.7% loss=0.052368 lr=5.626666e-06 elapsed=18:00:15 avg_step=1.33s recent_step=1.15s eta=18:57:55 +2026-08-25 21:56:01 | INFO | train rank=0 | step=48710/100000 progress=48.7% loss=0.053268 lr=5.625026e-06 elapsed=18:00:31 avg_step=1.33s recent_step=1.62s eta=18:57:45 +2026-08-25 21:56:13 | INFO | train rank=0 | step=48720/100000 progress=48.7% loss=0.041107 lr=5.623385e-06 elapsed=18:00:43 avg_step=1.33s recent_step=1.22s eta=18:57:30 +2026-08-25 21:56:29 | INFO | train rank=0 | step=48730/100000 progress=48.7% loss=0.123108 lr=5.621745e-06 elapsed=18:00:59 avg_step=1.33s recent_step=1.56s eta=18:57:19 +2026-08-25 21:56:44 | INFO | train rank=0 | step=48740/100000 progress=48.7% loss=0.083008 lr=5.620104e-06 elapsed=18:01:14 avg_step=1.33s recent_step=1.48s eta=18:57:08 +2026-08-25 21:56:55 | INFO | train rank=0 | step=48750/100000 progress=48.8% loss=0.082367 lr=5.618463e-06 elapsed=18:01:25 avg_step=1.33s recent_step=1.16s eta=18:56:53 +2026-08-25 21:57:09 | INFO | train rank=0 | step=48760/100000 progress=48.8% loss=0.058121 lr=5.616822e-06 elapsed=18:01:39 avg_step=1.33s recent_step=1.38s eta=18:56:40 +2026-08-25 21:57:22 | INFO | train rank=0 | step=48770/100000 progress=48.8% loss=0.034828 lr=5.615181e-06 elapsed=18:01:52 avg_step=1.33s recent_step=1.26s eta=18:56:26 +2026-08-25 21:57:39 | INFO | train rank=0 | step=48780/100000 progress=48.8% loss=0.062012 lr=5.613541e-06 elapsed=18:02:09 avg_step=1.33s recent_step=1.72s eta=18:56:17 +2026-08-25 21:57:53 | INFO | train rank=0 | step=48790/100000 progress=48.8% loss=0.085815 lr=5.611900e-06 elapsed=18:02:23 avg_step=1.33s recent_step=1.38s eta=18:56:04 +2026-08-25 21:58:07 | INFO | train rank=0 | step=48800/100000 progress=48.8% loss=0.063065 lr=5.610258e-06 elapsed=18:02:37 avg_step=1.33s recent_step=1.46s eta=18:55:52 +2026-08-25 21:58:20 | INFO | train rank=0 | step=48810/100000 progress=48.8% loss=0.048546 lr=5.608617e-06 elapsed=18:02:50 avg_step=1.33s recent_step=1.25s eta=18:55:38 +2026-08-25 21:58:33 | INFO | train rank=0 | step=48820/100000 progress=48.8% loss=0.055832 lr=5.606976e-06 elapsed=18:03:03 avg_step=1.33s recent_step=1.34s eta=18:55:24 +2026-08-25 21:58:45 | INFO | train rank=0 | step=48830/100000 progress=48.8% loss=0.069122 lr=5.605335e-06 elapsed=18:03:15 avg_step=1.33s recent_step=1.19s eta=18:55:10 +2026-08-25 21:58:57 | INFO | train rank=0 | step=48840/100000 progress=48.8% loss=0.059006 lr=5.603693e-06 elapsed=18:03:27 avg_step=1.33s recent_step=1.18s eta=18:54:55 +2026-08-25 21:59:11 | INFO | train rank=0 | step=48850/100000 progress=48.9% loss=0.048431 lr=5.602052e-06 elapsed=18:03:41 avg_step=1.33s recent_step=1.38s eta=18:54:42 +2026-08-25 21:59:27 | INFO | train rank=0 | step=48860/100000 progress=48.9% loss=0.037933 lr=5.600411e-06 elapsed=18:03:56 avg_step=1.33s recent_step=1.58s eta=18:54:31 +2026-08-25 21:59:44 | INFO | train rank=0 | step=48870/100000 progress=48.9% loss=0.035934 lr=5.598769e-06 elapsed=18:04:14 avg_step=1.33s recent_step=1.72s eta=18:54:22 +2026-08-25 21:59:57 | INFO | train rank=0 | step=48880/100000 progress=48.9% loss=0.066681 lr=5.597127e-06 elapsed=18:04:27 avg_step=1.33s recent_step=1.29s eta=18:54:08 +2026-08-25 22:00:08 | INFO | train rank=0 | step=48890/100000 progress=48.9% loss=0.084808 lr=5.595486e-06 elapsed=18:04:38 avg_step=1.33s recent_step=1.18s eta=18:53:53 +2026-08-25 22:00:21 | INFO | train rank=0 | step=48900/100000 progress=48.9% loss=0.070129 lr=5.593844e-06 elapsed=18:04:51 avg_step=1.33s recent_step=1.24s eta=18:53:39 +2026-08-25 22:00:33 | INFO | train rank=0 | step=48910/100000 progress=48.9% loss=0.033508 lr=5.592202e-06 elapsed=18:05:03 avg_step=1.33s recent_step=1.19s eta=18:53:24 +2026-08-25 22:00:45 | INFO | train rank=0 | step=48920/100000 progress=48.9% loss=0.073914 lr=5.590560e-06 elapsed=18:05:15 avg_step=1.33s recent_step=1.20s eta=18:53:10 +2026-08-25 22:00:58 | INFO | train rank=0 | step=48930/100000 progress=48.9% loss=0.097961 lr=5.588918e-06 elapsed=18:05:28 avg_step=1.33s recent_step=1.29s eta=18:52:56 +2026-08-25 22:01:13 | INFO | train rank=0 | step=48940/100000 progress=48.9% loss=0.060883 lr=5.587276e-06 elapsed=18:05:43 avg_step=1.33s recent_step=1.52s eta=18:52:45 +2026-08-25 22:01:24 | INFO | train rank=0 | step=48950/100000 progress=48.9% loss=0.064819 lr=5.585634e-06 elapsed=18:05:54 avg_step=1.33s recent_step=1.15s eta=18:52:29 +2026-08-25 22:01:38 | INFO | train rank=0 | step=48960/100000 progress=49.0% loss=0.064087 lr=5.583992e-06 elapsed=18:06:08 avg_step=1.33s recent_step=1.38s eta=18:52:17 +2026-08-25 22:01:50 | INFO | train rank=0 | step=48970/100000 progress=49.0% loss=0.052719 lr=5.582350e-06 elapsed=18:06:20 avg_step=1.33s recent_step=1.21s eta=18:52:02 +2026-08-25 22:02:06 | INFO | train rank=0 | step=48980/100000 progress=49.0% loss=0.045685 lr=5.580708e-06 elapsed=18:06:35 avg_step=1.33s recent_step=1.53s eta=18:51:51 +2026-08-25 22:02:17 | INFO | train rank=0 | step=48990/100000 progress=49.0% loss=0.144836 lr=5.579066e-06 elapsed=18:06:47 avg_step=1.33s recent_step=1.14s eta=18:51:35 +2026-08-25 22:02:30 | INFO | train rank=0 | step=49000/100000 progress=49.0% loss=0.039642 lr=5.577423e-06 elapsed=18:07:00 avg_step=1.33s recent_step=1.27s eta=18:51:22 +2026-08-25 22:02:44 | INFO | train rank=0 | step=49010/100000 progress=49.0% loss=0.061493 lr=5.575781e-06 elapsed=18:07:13 avg_step=1.33s recent_step=1.40s eta=18:51:09 +2026-08-25 22:02:56 | INFO | train rank=0 | step=49020/100000 progress=49.0% loss=0.059174 lr=5.574138e-06 elapsed=18:07:26 avg_step=1.33s recent_step=1.29s eta=18:50:55 +2026-08-25 22:03:13 | INFO | train rank=0 | step=49030/100000 progress=49.0% loss=0.050194 lr=5.572496e-06 elapsed=18:07:43 avg_step=1.33s recent_step=1.64s eta=18:50:45 +2026-08-25 22:03:24 | INFO | train rank=0 | step=49040/100000 progress=49.0% loss=0.066071 lr=5.570853e-06 elapsed=18:07:54 avg_step=1.33s recent_step=1.12s eta=18:50:30 +2026-08-25 22:03:36 | INFO | train rank=0 | step=49050/100000 progress=49.0% loss=0.100830 lr=5.569210e-06 elapsed=18:08:06 avg_step=1.33s recent_step=1.17s eta=18:50:15 +2026-08-25 22:03:48 | INFO | train rank=0 | step=49060/100000 progress=49.1% loss=0.060501 lr=5.567568e-06 elapsed=18:08:18 avg_step=1.33s recent_step=1.22s eta=18:50:00 +2026-08-25 22:04:04 | INFO | train rank=0 | step=49070/100000 progress=49.1% loss=0.080750 lr=5.565925e-06 elapsed=18:08:34 avg_step=1.33s recent_step=1.60s eta=18:49:50 +2026-08-25 22:04:18 | INFO | train rank=0 | step=49080/100000 progress=49.1% loss=0.053864 lr=5.564282e-06 elapsed=18:08:48 avg_step=1.33s recent_step=1.45s eta=18:49:37 +2026-08-25 22:04:32 | INFO | train rank=0 | step=49090/100000 progress=49.1% loss=0.072403 lr=5.562639e-06 elapsed=18:09:01 avg_step=1.33s recent_step=1.31s eta=18:49:24 +2026-08-25 22:04:45 | INFO | train rank=0 | step=49100/100000 progress=49.1% loss=0.073364 lr=5.560996e-06 elapsed=18:09:15 avg_step=1.33s recent_step=1.32s eta=18:49:11 +2026-08-25 22:05:00 | INFO | train rank=0 | step=49110/100000 progress=49.1% loss=0.030502 lr=5.559353e-06 elapsed=18:09:30 avg_step=1.33s recent_step=1.53s eta=18:48:59 +2026-08-25 22:05:11 | INFO | train rank=0 | step=49120/100000 progress=49.1% loss=0.083618 lr=5.557710e-06 elapsed=18:09:41 avg_step=1.33s recent_step=1.12s eta=18:48:44 +2026-08-25 22:05:27 | INFO | train rank=0 | step=49130/100000 progress=49.1% loss=0.067383 lr=5.556067e-06 elapsed=18:09:57 avg_step=1.33s recent_step=1.57s eta=18:48:33 +2026-08-25 22:05:40 | INFO | train rank=0 | step=49140/100000 progress=49.1% loss=0.047638 lr=5.554423e-06 elapsed=18:10:10 avg_step=1.33s recent_step=1.30s eta=18:48:19 +2026-08-25 22:05:55 | INFO | train rank=0 | step=49150/100000 progress=49.1% loss=0.052841 lr=5.552780e-06 elapsed=18:10:25 avg_step=1.33s recent_step=1.51s eta=18:48:08 +2026-08-25 22:06:07 | INFO | train rank=0 | step=49160/100000 progress=49.2% loss=0.059036 lr=5.551137e-06 elapsed=18:10:36 avg_step=1.33s recent_step=1.15s eta=18:47:53 +2026-08-25 22:06:19 | INFO | train rank=0 | step=49170/100000 progress=49.2% loss=0.042000 lr=5.549493e-06 elapsed=18:10:49 avg_step=1.33s recent_step=1.26s eta=18:47:39 +2026-08-25 22:06:36 | INFO | train rank=0 | step=49180/100000 progress=49.2% loss=0.053986 lr=5.547850e-06 elapsed=18:11:06 avg_step=1.33s recent_step=1.66s eta=18:47:29 +2026-08-25 22:06:49 | INFO | train rank=0 | step=49190/100000 progress=49.2% loss=0.041245 lr=5.546206e-06 elapsed=18:11:19 avg_step=1.33s recent_step=1.31s eta=18:47:15 +2026-08-25 22:07:00 | INFO | train rank=0 | step=49200/100000 progress=49.2% loss=0.056274 lr=5.544563e-06 elapsed=18:11:30 avg_step=1.33s recent_step=1.14s eta=18:47:00 +2026-08-25 22:07:15 | INFO | train rank=0 | step=49210/100000 progress=49.2% loss=0.043633 lr=5.542919e-06 elapsed=18:11:45 avg_step=1.33s recent_step=1.49s eta=18:46:48 +2026-08-25 22:07:30 | INFO | train rank=0 | step=49220/100000 progress=49.2% loss=0.064087 lr=5.541275e-06 elapsed=18:12:00 avg_step=1.33s recent_step=1.53s eta=18:46:37 +2026-08-25 22:07:45 | INFO | train rank=0 | step=49230/100000 progress=49.2% loss=0.040863 lr=5.539631e-06 elapsed=18:12:15 avg_step=1.33s recent_step=1.43s eta=18:46:25 +2026-08-25 22:07:57 | INFO | train rank=0 | step=49240/100000 progress=49.2% loss=0.057693 lr=5.537988e-06 elapsed=18:12:26 avg_step=1.33s recent_step=1.19s eta=18:46:10 +2026-08-25 22:08:13 | INFO | train rank=0 | step=49250/100000 progress=49.2% loss=0.028091 lr=5.536344e-06 elapsed=18:12:43 avg_step=1.33s recent_step=1.69s eta=18:46:00 +2026-08-25 22:08:25 | INFO | train rank=0 | step=49260/100000 progress=49.3% loss=0.071640 lr=5.534700e-06 elapsed=18:12:55 avg_step=1.33s recent_step=1.15s eta=18:45:45 +2026-08-25 22:08:40 | INFO | train rank=0 | step=49270/100000 progress=49.3% loss=0.150269 lr=5.533056e-06 elapsed=18:13:10 avg_step=1.33s recent_step=1.48s eta=18:45:33 +2026-08-25 22:08:52 | INFO | train rank=0 | step=49280/100000 progress=49.3% loss=0.063385 lr=5.531412e-06 elapsed=18:13:21 avg_step=1.33s recent_step=1.19s eta=18:45:18 +2026-08-25 22:09:05 | INFO | train rank=0 | step=49290/100000 progress=49.3% loss=0.061707 lr=5.529768e-06 elapsed=18:13:35 avg_step=1.33s recent_step=1.37s eta=18:45:05 +2026-08-25 22:09:20 | INFO | train rank=0 | step=49300/100000 progress=49.3% loss=0.044022 lr=5.528123e-06 elapsed=18:13:50 avg_step=1.33s recent_step=1.48s eta=18:44:54 +2026-08-25 22:09:35 | INFO | train rank=0 | step=49310/100000 progress=49.3% loss=0.044128 lr=5.526479e-06 elapsed=18:14:05 avg_step=1.33s recent_step=1.48s eta=18:44:42 +2026-08-25 22:09:47 | INFO | train rank=0 | step=49320/100000 progress=49.3% loss=0.067047 lr=5.524835e-06 elapsed=18:14:16 avg_step=1.33s recent_step=1.18s eta=18:44:27 +2026-08-25 22:10:00 | INFO | train rank=0 | step=49330/100000 progress=49.3% loss=0.041763 lr=5.523190e-06 elapsed=18:14:30 avg_step=1.33s recent_step=1.39s eta=18:44:14 +2026-08-25 22:10:12 | INFO | train rank=0 | step=49340/100000 progress=49.3% loss=0.065430 lr=5.521546e-06 elapsed=18:14:42 avg_step=1.33s recent_step=1.15s eta=18:43:59 +2026-08-25 22:10:28 | INFO | train rank=0 | step=49350/100000 progress=49.4% loss=0.057266 lr=5.519902e-06 elapsed=18:14:58 avg_step=1.33s recent_step=1.58s eta=18:43:48 +2026-08-25 22:10:41 | INFO | train rank=0 | step=49360/100000 progress=49.4% loss=0.047928 lr=5.518257e-06 elapsed=18:15:11 avg_step=1.33s recent_step=1.32s eta=18:43:35 +2026-08-25 22:10:54 | INFO | train rank=0 | step=49370/100000 progress=49.4% loss=0.036888 lr=5.516612e-06 elapsed=18:15:24 avg_step=1.33s recent_step=1.29s eta=18:43:21 +2026-08-25 22:11:11 | INFO | train rank=0 | step=49380/100000 progress=49.4% loss=0.031731 lr=5.514968e-06 elapsed=18:15:41 avg_step=1.33s recent_step=1.71s eta=18:43:12 +2026-08-25 22:11:23 | INFO | train rank=0 | step=49390/100000 progress=49.4% loss=0.050507 lr=5.513323e-06 elapsed=18:15:53 avg_step=1.33s recent_step=1.18s eta=18:42:57 +2026-08-25 22:11:37 | INFO | train rank=0 | step=49400/100000 progress=49.4% loss=0.050049 lr=5.511678e-06 elapsed=18:16:07 avg_step=1.33s recent_step=1.44s eta=18:42:45 +2026-08-25 22:11:48 | INFO | train rank=0 | step=49410/100000 progress=49.4% loss=0.046036 lr=5.510033e-06 elapsed=18:16:18 avg_step=1.33s recent_step=1.12s eta=18:42:29 +2026-08-25 22:12:00 | INFO | train rank=0 | step=49420/100000 progress=49.4% loss=0.042664 lr=5.508389e-06 elapsed=18:16:30 avg_step=1.33s recent_step=1.22s eta=18:42:15 +2026-08-25 22:12:12 | INFO | train rank=0 | step=49430/100000 progress=49.4% loss=0.042221 lr=5.506744e-06 elapsed=18:16:42 avg_step=1.33s recent_step=1.16s eta=18:42:00 +2026-08-25 22:12:23 | INFO | train rank=0 | step=49440/100000 progress=49.4% loss=0.061157 lr=5.505099e-06 elapsed=18:16:53 avg_step=1.33s recent_step=1.14s eta=18:41:44 +2026-08-25 22:12:37 | INFO | train rank=0 | step=49450/100000 progress=49.5% loss=0.058945 lr=5.503454e-06 elapsed=18:17:07 avg_step=1.33s recent_step=1.39s eta=18:41:32 +2026-08-25 22:12:49 | INFO | train rank=0 | step=49460/100000 progress=49.5% loss=0.049301 lr=5.501809e-06 elapsed=18:17:19 avg_step=1.33s recent_step=1.15s eta=18:41:16 +2026-08-25 22:13:03 | INFO | train rank=0 | step=49470/100000 progress=49.5% loss=0.054565 lr=5.500163e-06 elapsed=18:17:33 avg_step=1.33s recent_step=1.42s eta=18:41:04 +2026-08-25 22:13:18 | INFO | train rank=0 | step=49480/100000 progress=49.5% loss=0.040359 lr=5.498518e-06 elapsed=18:17:48 avg_step=1.33s recent_step=1.49s eta=18:40:52 +2026-08-25 22:13:31 | INFO | train rank=0 | step=49490/100000 progress=49.5% loss=0.086121 lr=5.496873e-06 elapsed=18:18:01 avg_step=1.33s recent_step=1.28s eta=18:40:39 +2026-08-25 22:13:48 | INFO | train rank=0 | step=49500/100000 progress=49.5% loss=0.029137 lr=5.495228e-06 elapsed=18:18:18 avg_step=1.33s recent_step=1.69s eta=18:40:29 +2026-08-25 22:14:01 | INFO | train rank=0 | step=49510/100000 progress=49.5% loss=0.052902 lr=5.493582e-06 elapsed=18:18:31 avg_step=1.33s recent_step=1.36s eta=18:40:16 +2026-08-25 22:14:13 | INFO | train rank=0 | step=49520/100000 progress=49.5% loss=0.028519 lr=5.491937e-06 elapsed=18:18:43 avg_step=1.33s recent_step=1.18s eta=18:40:01 +2026-08-25 22:14:25 | INFO | train rank=0 | step=49530/100000 progress=49.5% loss=0.053513 lr=5.490291e-06 elapsed=18:18:55 avg_step=1.33s recent_step=1.22s eta=18:39:47 +2026-08-25 22:14:40 | INFO | train rank=0 | step=49540/100000 progress=49.5% loss=0.070038 lr=5.488646e-06 elapsed=18:19:10 avg_step=1.33s recent_step=1.48s eta=18:39:35 +2026-08-25 22:14:58 | INFO | train rank=0 | step=49550/100000 progress=49.5% loss=0.082092 lr=5.487000e-06 elapsed=18:19:28 avg_step=1.33s recent_step=1.82s eta=18:39:26 +2026-08-25 22:15:10 | INFO | train rank=0 | step=49560/100000 progress=49.6% loss=0.044342 lr=5.485355e-06 elapsed=18:19:39 avg_step=1.33s recent_step=1.13s eta=18:39:11 +2026-08-25 22:15:21 | INFO | train rank=0 | step=49570/100000 progress=49.6% loss=0.043488 lr=5.483709e-06 elapsed=18:19:51 avg_step=1.33s recent_step=1.18s eta=18:38:56 +2026-08-25 22:15:34 | INFO | train rank=0 | step=49580/100000 progress=49.6% loss=0.043518 lr=5.482063e-06 elapsed=18:20:04 avg_step=1.33s recent_step=1.26s eta=18:38:42 +2026-08-25 22:15:47 | INFO | train rank=0 | step=49590/100000 progress=49.6% loss=0.063187 lr=5.480417e-06 elapsed=18:20:17 avg_step=1.33s recent_step=1.31s eta=18:38:29 +2026-08-25 22:16:02 | INFO | train rank=0 | step=49600/100000 progress=49.6% loss=0.057526 lr=5.478772e-06 elapsed=18:20:32 avg_step=1.33s recent_step=1.46s eta=18:38:17 +2026-08-25 22:16:13 | INFO | train rank=0 | step=49610/100000 progress=49.6% loss=0.066193 lr=5.477126e-06 elapsed=18:20:43 avg_step=1.33s recent_step=1.15s eta=18:38:01 +2026-08-25 22:16:29 | INFO | train rank=0 | step=49620/100000 progress=49.6% loss=0.046936 lr=5.475480e-06 elapsed=18:20:59 avg_step=1.33s recent_step=1.55s eta=18:37:50 +2026-08-25 22:16:45 | INFO | train rank=0 | step=49630/100000 progress=49.6% loss=0.049576 lr=5.473834e-06 elapsed=18:21:15 avg_step=1.33s recent_step=1.66s eta=18:37:40 +2026-08-25 22:16:57 | INFO | train rank=0 | step=49640/100000 progress=49.6% loss=0.056778 lr=5.472188e-06 elapsed=18:21:26 avg_step=1.33s recent_step=1.13s eta=18:37:25 +2026-08-25 22:17:08 | INFO | train rank=0 | step=49650/100000 progress=49.6% loss=0.058640 lr=5.470542e-06 elapsed=18:21:38 avg_step=1.33s recent_step=1.14s eta=18:37:10 +2026-08-25 22:17:24 | INFO | train rank=0 | step=49660/100000 progress=49.7% loss=0.044846 lr=5.468895e-06 elapsed=18:21:53 avg_step=1.33s recent_step=1.56s eta=18:36:59 +2026-08-25 22:17:35 | INFO | train rank=0 | step=49670/100000 progress=49.7% loss=0.113403 lr=5.467249e-06 elapsed=18:22:05 avg_step=1.33s recent_step=1.14s eta=18:36:43 +2026-08-25 22:17:47 | INFO | train rank=0 | step=49680/100000 progress=49.7% loss=0.056259 lr=5.465603e-06 elapsed=18:22:16 avg_step=1.33s recent_step=1.16s eta=18:36:28 +2026-08-25 22:17:58 | INFO | train rank=0 | step=49690/100000 progress=49.7% loss=0.052132 lr=5.463957e-06 elapsed=18:22:28 avg_step=1.33s recent_step=1.14s eta=18:36:13 +2026-08-25 22:18:13 | INFO | train rank=0 | step=49700/100000 progress=49.7% loss=0.035248 lr=5.462310e-06 elapsed=18:22:43 avg_step=1.33s recent_step=1.50s eta=18:36:02 +2026-08-25 22:18:28 | INFO | train rank=0 | step=49710/100000 progress=49.7% loss=0.050232 lr=5.460664e-06 elapsed=18:22:57 avg_step=1.33s recent_step=1.46s eta=18:35:50 +2026-08-25 22:18:39 | INFO | train rank=0 | step=49720/100000 progress=49.7% loss=0.039299 lr=5.459017e-06 elapsed=18:23:09 avg_step=1.33s recent_step=1.13s eta=18:35:34 +2026-08-25 22:18:50 | INFO | train rank=0 | step=49730/100000 progress=49.7% loss=0.053802 lr=5.457371e-06 elapsed=18:23:20 avg_step=1.33s recent_step=1.16s eta=18:35:19 +2026-08-25 22:19:02 | INFO | train rank=0 | step=49740/100000 progress=49.7% loss=0.057892 lr=5.455724e-06 elapsed=18:23:32 avg_step=1.33s recent_step=1.18s eta=18:35:04 +2026-08-25 22:19:16 | INFO | train rank=0 | step=49750/100000 progress=49.8% loss=0.076263 lr=5.454078e-06 elapsed=18:23:46 avg_step=1.33s recent_step=1.35s eta=18:34:51 +2026-08-25 22:19:28 | INFO | train rank=0 | step=49760/100000 progress=49.8% loss=0.087585 lr=5.452431e-06 elapsed=18:23:58 avg_step=1.33s recent_step=1.18s eta=18:34:36 +2026-08-25 22:19:40 | INFO | train rank=0 | step=49770/100000 progress=49.8% loss=0.055634 lr=5.450784e-06 elapsed=18:24:10 avg_step=1.33s recent_step=1.21s eta=18:34:22 +2026-08-25 22:19:53 | INFO | train rank=0 | step=49780/100000 progress=49.8% loss=0.070526 lr=5.449138e-06 elapsed=18:24:23 avg_step=1.33s recent_step=1.32s eta=18:34:08 +2026-08-25 22:20:07 | INFO | train rank=0 | step=49790/100000 progress=49.8% loss=0.045319 lr=5.447491e-06 elapsed=18:24:37 avg_step=1.33s recent_step=1.38s eta=18:33:56 +2026-08-25 22:20:20 | INFO | train rank=0 | step=49800/100000 progress=49.8% loss=0.040802 lr=5.445844e-06 elapsed=18:24:50 avg_step=1.33s recent_step=1.33s eta=18:33:42 +2026-08-25 22:20:34 | INFO | train rank=0 | step=49810/100000 progress=49.8% loss=0.041931 lr=5.444197e-06 elapsed=18:25:04 avg_step=1.33s recent_step=1.43s eta=18:33:30 +2026-08-25 22:20:49 | INFO | train rank=0 | step=49820/100000 progress=49.8% loss=0.050903 lr=5.442550e-06 elapsed=18:25:19 avg_step=1.33s recent_step=1.45s eta=18:33:18 +2026-08-25 22:21:00 | INFO | train rank=0 | step=49830/100000 progress=49.8% loss=0.050934 lr=5.440903e-06 elapsed=18:25:30 avg_step=1.33s recent_step=1.15s eta=18:33:03 +2026-08-25 22:21:17 | INFO | train rank=0 | step=49840/100000 progress=49.8% loss=0.029015 lr=5.439256e-06 elapsed=18:25:47 avg_step=1.33s recent_step=1.64s eta=18:32:53 +2026-08-25 22:21:29 | INFO | train rank=0 | step=49850/100000 progress=49.9% loss=0.036743 lr=5.437609e-06 elapsed=18:25:59 avg_step=1.33s recent_step=1.25s eta=18:32:38 +2026-08-25 22:21:41 | INFO | train rank=0 | step=49860/100000 progress=49.9% loss=0.071671 lr=5.435962e-06 elapsed=18:26:11 avg_step=1.33s recent_step=1.17s eta=18:32:23 +2026-08-25 22:21:56 | INFO | train rank=0 | step=49870/100000 progress=49.9% loss=0.076172 lr=5.434315e-06 elapsed=18:26:26 avg_step=1.33s recent_step=1.49s eta=18:32:12 +2026-08-25 22:22:10 | INFO | train rank=0 | step=49880/100000 progress=49.9% loss=0.050476 lr=5.432667e-06 elapsed=18:26:40 avg_step=1.33s recent_step=1.42s eta=18:31:59 +2026-08-25 22:22:27 | INFO | train rank=0 | step=49890/100000 progress=49.9% loss=0.049225 lr=5.431020e-06 elapsed=18:26:57 avg_step=1.33s recent_step=1.69s eta=18:31:50 +2026-08-25 22:22:38 | INFO | train rank=0 | step=49900/100000 progress=49.9% loss=0.074280 lr=5.429373e-06 elapsed=18:27:08 avg_step=1.33s recent_step=1.14s eta=18:31:34 +2026-08-25 22:22:54 | INFO | train rank=0 | step=49910/100000 progress=49.9% loss=0.040070 lr=5.427725e-06 elapsed=18:27:24 avg_step=1.33s recent_step=1.55s eta=18:31:23 +2026-08-25 22:23:07 | INFO | train rank=0 | step=49920/100000 progress=49.9% loss=0.054871 lr=5.426078e-06 elapsed=18:27:37 avg_step=1.33s recent_step=1.32s eta=18:31:10 +2026-08-25 22:23:19 | INFO | train rank=0 | step=49930/100000 progress=49.9% loss=0.050491 lr=5.424430e-06 elapsed=18:27:49 avg_step=1.33s recent_step=1.25s eta=18:30:56 +2026-08-25 22:23:33 | INFO | train rank=0 | step=49940/100000 progress=49.9% loss=0.048309 lr=5.422783e-06 elapsed=18:28:03 avg_step=1.33s recent_step=1.33s eta=18:30:42 +2026-08-25 22:23:44 | INFO | train rank=0 | step=49950/100000 progress=50.0% loss=0.030075 lr=5.421135e-06 elapsed=18:28:14 avg_step=1.33s recent_step=1.17s eta=18:30:27 +2026-08-25 22:24:00 | INFO | train rank=0 | step=49960/100000 progress=50.0% loss=0.036537 lr=5.419488e-06 elapsed=18:28:30 avg_step=1.33s recent_step=1.58s eta=18:30:17 +2026-08-25 22:24:12 | INFO | train rank=0 | step=49970/100000 progress=50.0% loss=0.047760 lr=5.417840e-06 elapsed=18:28:42 avg_step=1.33s recent_step=1.18s eta=18:30:02 +2026-08-25 22:24:24 | INFO | train rank=0 | step=49980/100000 progress=50.0% loss=0.063416 lr=5.416192e-06 elapsed=18:28:54 avg_step=1.33s recent_step=1.18s eta=18:29:47 +2026-08-25 22:24:37 | INFO | train rank=0 | step=49990/100000 progress=50.0% loss=0.051285 lr=5.414545e-06 elapsed=18:29:07 avg_step=1.33s recent_step=1.32s eta=18:29:34 +2026-08-25 22:24:53 | INFO | train rank=0 | step=50000/100000 progress=50.0% loss=0.058792 lr=5.412897e-06 elapsed=18:29:23 avg_step=1.33s recent_step=1.59s eta=18:29:23 +2026-08-25 22:25:06 | INFO | train rank=0 | step=50010/100000 progress=50.0% loss=0.047211 lr=5.411249e-06 elapsed=18:29:36 avg_step=1.33s recent_step=1.34s eta=18:29:10 +2026-08-25 22:25:19 | INFO | train rank=0 | step=50020/100000 progress=50.0% loss=0.059692 lr=5.409601e-06 elapsed=18:29:48 avg_step=1.33s recent_step=1.22s eta=18:28:55 +2026-08-25 22:25:34 | INFO | train rank=0 | step=50030/100000 progress=50.0% loss=0.047089 lr=5.407953e-06 elapsed=18:30:04 avg_step=1.33s recent_step=1.51s eta=18:28:44 +2026-08-25 22:25:46 | INFO | train rank=0 | step=50040/100000 progress=50.0% loss=0.041412 lr=5.406305e-06 elapsed=18:30:16 avg_step=1.33s recent_step=1.23s eta=18:28:29 +2026-08-25 22:26:01 | INFO | train rank=0 | step=50050/100000 progress=50.0% loss=0.035416 lr=5.404657e-06 elapsed=18:30:31 avg_step=1.33s recent_step=1.49s eta=18:28:18 +2026-08-25 22:26:13 | INFO | train rank=0 | step=50060/100000 progress=50.1% loss=0.059753 lr=5.403009e-06 elapsed=18:30:43 avg_step=1.33s recent_step=1.22s eta=18:28:03 +2026-08-25 22:26:27 | INFO | train rank=0 | step=50070/100000 progress=50.1% loss=0.036026 lr=5.401361e-06 elapsed=18:30:57 avg_step=1.33s recent_step=1.38s eta=18:27:50 +2026-08-25 22:26:41 | INFO | train rank=0 | step=50080/100000 progress=50.1% loss=0.057175 lr=5.399713e-06 elapsed=18:31:11 avg_step=1.33s recent_step=1.46s eta=18:27:38 +2026-08-25 22:26:59 | INFO | train rank=0 | step=50090/100000 progress=50.1% loss=0.046623 lr=5.398065e-06 elapsed=18:31:29 avg_step=1.33s recent_step=1.77s eta=18:27:29 +2026-08-25 22:27:13 | INFO | train rank=0 | step=50100/100000 progress=50.1% loss=0.044586 lr=5.396416e-06 elapsed=18:31:43 avg_step=1.33s recent_step=1.37s eta=18:27:17 +2026-08-25 22:27:25 | INFO | train rank=0 | step=50110/100000 progress=50.1% loss=0.040527 lr=5.394768e-06 elapsed=18:31:55 avg_step=1.33s recent_step=1.19s eta=18:27:02 +2026-08-25 22:27:40 | INFO | train rank=0 | step=50120/100000 progress=50.1% loss=0.051849 lr=5.393120e-06 elapsed=18:32:10 avg_step=1.33s recent_step=1.48s eta=18:26:50 +2026-08-25 22:27:53 | INFO | train rank=0 | step=50130/100000 progress=50.1% loss=0.036751 lr=5.391471e-06 elapsed=18:32:23 avg_step=1.33s recent_step=1.32s eta=18:26:37 +2026-08-25 22:28:07 | INFO | train rank=0 | step=50140/100000 progress=50.1% loss=0.047867 lr=5.389823e-06 elapsed=18:32:37 avg_step=1.33s recent_step=1.40s eta=18:26:24 +2026-08-25 22:28:18 | INFO | train rank=0 | step=50150/100000 progress=50.1% loss=0.055145 lr=5.388174e-06 elapsed=18:32:48 avg_step=1.33s recent_step=1.15s eta=18:26:09 +2026-08-25 22:28:30 | INFO | train rank=0 | step=50160/100000 progress=50.2% loss=0.066071 lr=5.386526e-06 elapsed=18:33:00 avg_step=1.33s recent_step=1.18s eta=18:25:54 +2026-08-25 22:28:45 | INFO | train rank=0 | step=50170/100000 progress=50.2% loss=0.035934 lr=5.384877e-06 elapsed=18:33:15 avg_step=1.33s recent_step=1.47s eta=18:25:42 +2026-08-25 22:28:59 | INFO | train rank=0 | step=50180/100000 progress=50.2% loss=0.050903 lr=5.383229e-06 elapsed=18:33:29 avg_step=1.33s recent_step=1.41s eta=18:25:30 +2026-08-25 22:29:14 | INFO | train rank=0 | step=50190/100000 progress=50.2% loss=0.045715 lr=5.381580e-06 elapsed=18:33:44 avg_step=1.33s recent_step=1.54s eta=18:25:18 +2026-08-25 22:29:29 | INFO | train rank=0 | step=50200/100000 progress=50.2% loss=0.047913 lr=5.379931e-06 elapsed=18:33:59 avg_step=1.33s recent_step=1.42s eta=18:25:06 +2026-08-25 22:29:44 | INFO | train rank=0 | step=50210/100000 progress=50.2% loss=0.099609 lr=5.378283e-06 elapsed=18:34:14 avg_step=1.33s recent_step=1.54s eta=18:24:55 +2026-08-25 22:29:58 | INFO | train rank=0 | step=50220/100000 progress=50.2% loss=0.049103 lr=5.376634e-06 elapsed=18:34:28 avg_step=1.33s recent_step=1.42s eta=18:24:42 +2026-08-25 22:30:11 | INFO | train rank=0 | step=50230/100000 progress=50.2% loss=0.079010 lr=5.374985e-06 elapsed=18:34:41 avg_step=1.33s recent_step=1.33s eta=18:24:29 +2026-08-25 22:30:24 | INFO | train rank=0 | step=50240/100000 progress=50.2% loss=0.045654 lr=5.373336e-06 elapsed=18:34:53 avg_step=1.33s recent_step=1.21s eta=18:24:14 +2026-08-25 22:30:36 | INFO | train rank=0 | step=50250/100000 progress=50.2% loss=0.029778 lr=5.371688e-06 elapsed=18:35:06 avg_step=1.33s recent_step=1.21s eta=18:24:00 +2026-08-25 22:30:51 | INFO | train rank=0 | step=50260/100000 progress=50.3% loss=0.041824 lr=5.370039e-06 elapsed=18:35:21 avg_step=1.33s recent_step=1.53s eta=18:23:49 +2026-08-25 22:31:03 | INFO | train rank=0 | step=50270/100000 progress=50.3% loss=0.039185 lr=5.368390e-06 elapsed=18:35:33 avg_step=1.33s recent_step=1.19s eta=18:23:34 +2026-08-25 22:31:19 | INFO | train rank=0 | step=50280/100000 progress=50.3% loss=0.055344 lr=5.366741e-06 elapsed=18:35:49 avg_step=1.33s recent_step=1.60s eta=18:23:23 +2026-08-25 22:31:30 | INFO | train rank=0 | step=50290/100000 progress=50.3% loss=0.061493 lr=5.365092e-06 elapsed=18:36:00 avg_step=1.33s recent_step=1.15s eta=18:23:08 +2026-08-25 22:31:43 | INFO | train rank=0 | step=50300/100000 progress=50.3% loss=0.071594 lr=5.363443e-06 elapsed=18:36:13 avg_step=1.33s recent_step=1.22s eta=18:22:54 +2026-08-25 22:31:57 | INFO | train rank=0 | step=50310/100000 progress=50.3% loss=0.053162 lr=5.361793e-06 elapsed=18:36:27 avg_step=1.33s recent_step=1.41s eta=18:22:41 +2026-08-25 22:32:09 | INFO | train rank=0 | step=50320/100000 progress=50.3% loss=0.054688 lr=5.360144e-06 elapsed=18:36:39 avg_step=1.33s recent_step=1.19s eta=18:22:27 +2026-08-25 22:32:22 | INFO | train rank=0 | step=50330/100000 progress=50.3% loss=0.071930 lr=5.358495e-06 elapsed=18:36:52 avg_step=1.33s recent_step=1.36s eta=18:22:14 +2026-08-25 22:32:34 | INFO | train rank=0 | step=50340/100000 progress=50.3% loss=0.050247 lr=5.356846e-06 elapsed=18:37:04 avg_step=1.33s recent_step=1.17s eta=18:21:59 +2026-08-25 22:32:49 | INFO | train rank=0 | step=50350/100000 progress=50.3% loss=0.027824 lr=5.355197e-06 elapsed=18:37:18 avg_step=1.33s recent_step=1.45s eta=18:21:46 +2026-08-25 22:33:03 | INFO | train rank=0 | step=50360/100000 progress=50.4% loss=0.074142 lr=5.353547e-06 elapsed=18:37:33 avg_step=1.33s recent_step=1.42s eta=18:21:34 +2026-08-25 22:33:14 | INFO | train rank=0 | step=50370/100000 progress=50.4% loss=0.060638 lr=5.351898e-06 elapsed=18:37:44 avg_step=1.33s recent_step=1.14s eta=18:21:19 +2026-08-25 22:33:26 | INFO | train rank=0 | step=50380/100000 progress=50.4% loss=0.043869 lr=5.350249e-06 elapsed=18:37:56 avg_step=1.33s recent_step=1.19s eta=18:21:04 +2026-08-25 22:33:38 | INFO | train rank=0 | step=50390/100000 progress=50.4% loss=0.102295 lr=5.348599e-06 elapsed=18:38:08 avg_step=1.33s recent_step=1.16s eta=18:20:49 +2026-08-25 22:33:49 | INFO | train rank=0 | step=50400/100000 progress=50.4% loss=0.055878 lr=5.346950e-06 elapsed=18:38:19 avg_step=1.33s recent_step=1.17s eta=18:20:34 +2026-08-25 22:34:04 | INFO | train rank=0 | step=50410/100000 progress=50.4% loss=0.111816 lr=5.345300e-06 elapsed=18:38:34 avg_step=1.33s recent_step=1.50s eta=18:20:23 +2026-08-25 22:34:17 | INFO | train rank=0 | step=50420/100000 progress=50.4% loss=0.034988 lr=5.343651e-06 elapsed=18:38:46 avg_step=1.33s recent_step=1.22s eta=18:20:08 +2026-08-25 22:34:32 | INFO | train rank=0 | step=50430/100000 progress=50.4% loss=0.041199 lr=5.342001e-06 elapsed=18:39:02 avg_step=1.33s recent_step=1.55s eta=18:19:57 +2026-08-25 22:34:44 | INFO | train rank=0 | step=50440/100000 progress=50.4% loss=0.035973 lr=5.340351e-06 elapsed=18:39:14 avg_step=1.33s recent_step=1.19s eta=18:19:42 +2026-08-25 22:35:00 | INFO | train rank=0 | step=50450/100000 progress=50.4% loss=0.038803 lr=5.338702e-06 elapsed=18:39:29 avg_step=1.33s recent_step=1.56s eta=18:19:31 +2026-08-25 22:35:12 | INFO | train rank=0 | step=50460/100000 progress=50.5% loss=0.044846 lr=5.337052e-06 elapsed=18:39:42 avg_step=1.33s recent_step=1.21s eta=18:19:17 +2026-08-25 22:35:26 | INFO | train rank=0 | step=50470/100000 progress=50.5% loss=0.059967 lr=5.335402e-06 elapsed=18:39:56 avg_step=1.33s recent_step=1.41s eta=18:19:04 +2026-08-25 22:35:37 | INFO | train rank=0 | step=50480/100000 progress=50.5% loss=0.067932 lr=5.333753e-06 elapsed=18:40:07 avg_step=1.33s recent_step=1.14s eta=18:18:49 +2026-08-25 22:35:55 | INFO | train rank=0 | step=50490/100000 progress=50.5% loss=0.069000 lr=5.332103e-06 elapsed=18:40:25 avg_step=1.33s recent_step=1.79s eta=18:18:40 +2026-08-25 22:36:07 | INFO | train rank=0 | step=50500/100000 progress=50.5% loss=0.099365 lr=5.330453e-06 elapsed=18:40:37 avg_step=1.33s recent_step=1.16s eta=18:18:25 +2026-08-25 22:36:19 | INFO | train rank=0 | step=50510/100000 progress=50.5% loss=0.072205 lr=5.328803e-06 elapsed=18:40:49 avg_step=1.33s recent_step=1.22s eta=18:18:11 +2026-08-25 22:36:35 | INFO | train rank=0 | step=50520/100000 progress=50.5% loss=0.089478 lr=5.327153e-06 elapsed=18:41:05 avg_step=1.33s recent_step=1.63s eta=18:18:00 +2026-08-25 22:36:47 | INFO | train rank=0 | step=50530/100000 progress=50.5% loss=0.041428 lr=5.325503e-06 elapsed=18:41:17 avg_step=1.33s recent_step=1.16s eta=18:17:45 +2026-08-25 22:37:02 | INFO | train rank=0 | step=50540/100000 progress=50.5% loss=0.036087 lr=5.323853e-06 elapsed=18:41:32 avg_step=1.33s recent_step=1.49s eta=18:17:34 +2026-08-25 22:37:13 | INFO | train rank=0 | step=50550/100000 progress=50.5% loss=0.040001 lr=5.322203e-06 elapsed=18:41:43 avg_step=1.33s recent_step=1.16s eta=18:17:19 +2026-08-25 22:37:27 | INFO | train rank=0 | step=50560/100000 progress=50.6% loss=0.043175 lr=5.320553e-06 elapsed=18:41:57 avg_step=1.33s recent_step=1.41s eta=18:17:06 +2026-08-25 22:37:44 | INFO | train rank=0 | step=50570/100000 progress=50.6% loss=0.069931 lr=5.318903e-06 elapsed=18:42:14 avg_step=1.33s recent_step=1.63s eta=18:16:56 +2026-08-25 22:37:55 | INFO | train rank=0 | step=50580/100000 progress=50.6% loss=0.036224 lr=5.317253e-06 elapsed=18:42:25 avg_step=1.33s recent_step=1.16s eta=18:16:41 +2026-08-25 22:38:11 | INFO | train rank=0 | step=50590/100000 progress=50.6% loss=0.054749 lr=5.315603e-06 elapsed=18:42:41 avg_step=1.33s recent_step=1.55s eta=18:16:29 +2026-08-25 22:38:23 | INFO | train rank=0 | step=50600/100000 progress=50.6% loss=0.051239 lr=5.313953e-06 elapsed=18:42:53 avg_step=1.33s recent_step=1.24s eta=18:16:15 +2026-08-25 22:38:35 | INFO | train rank=0 | step=50610/100000 progress=50.6% loss=0.071472 lr=5.312302e-06 elapsed=18:43:05 avg_step=1.33s recent_step=1.20s eta=18:16:01 +2026-08-25 22:38:50 | INFO | train rank=0 | step=50620/100000 progress=50.6% loss=0.027855 lr=5.310652e-06 elapsed=18:43:19 avg_step=1.33s recent_step=1.43s eta=18:15:48 +2026-08-25 22:39:05 | INFO | train rank=0 | step=50630/100000 progress=50.6% loss=0.038895 lr=5.309002e-06 elapsed=18:43:35 avg_step=1.33s recent_step=1.52s eta=18:15:37 +2026-08-25 22:39:19 | INFO | train rank=0 | step=50640/100000 progress=50.6% loss=0.059235 lr=5.307351e-06 elapsed=18:43:48 avg_step=1.33s recent_step=1.38s eta=18:15:24 +2026-08-25 22:39:31 | INFO | train rank=0 | step=50650/100000 progress=50.6% loss=0.046944 lr=5.305701e-06 elapsed=18:44:01 avg_step=1.33s recent_step=1.22s eta=18:15:10 +2026-08-25 22:39:45 | INFO | train rank=0 | step=50660/100000 progress=50.7% loss=0.030579 lr=5.304051e-06 elapsed=18:44:15 avg_step=1.33s recent_step=1.45s eta=18:14:57 +2026-08-25 22:39:57 | INFO | train rank=0 | step=50670/100000 progress=50.7% loss=0.040787 lr=5.302400e-06 elapsed=18:44:27 avg_step=1.33s recent_step=1.21s eta=18:14:43 +2026-08-25 22:40:09 | INFO | train rank=0 | step=50680/100000 progress=50.7% loss=0.086548 lr=5.300750e-06 elapsed=18:44:39 avg_step=1.33s recent_step=1.16s eta=18:14:28 +2026-08-25 22:40:22 | INFO | train rank=0 | step=50690/100000 progress=50.7% loss=0.056458 lr=5.299099e-06 elapsed=18:44:52 avg_step=1.33s recent_step=1.32s eta=18:14:15 +2026-08-25 22:40:34 | INFO | train rank=0 | step=50700/100000 progress=50.7% loss=0.028595 lr=5.297449e-06 elapsed=18:45:04 avg_step=1.33s recent_step=1.17s eta=18:14:00 +2026-08-25 22:40:49 | INFO | train rank=0 | step=50710/100000 progress=50.7% loss=0.042099 lr=5.295798e-06 elapsed=18:45:19 avg_step=1.33s recent_step=1.52s eta=18:13:48 +2026-08-25 22:41:01 | INFO | train rank=0 | step=50720/100000 progress=50.7% loss=0.042191 lr=5.294148e-06 elapsed=18:45:31 avg_step=1.33s recent_step=1.20s eta=18:13:34 +2026-08-25 22:41:16 | INFO | train rank=0 | step=50730/100000 progress=50.7% loss=0.052673 lr=5.292497e-06 elapsed=18:45:45 avg_step=1.33s recent_step=1.44s eta=18:13:21 +2026-08-25 22:41:29 | INFO | train rank=0 | step=50740/100000 progress=50.7% loss=0.132629 lr=5.290846e-06 elapsed=18:45:59 avg_step=1.33s recent_step=1.34s eta=18:13:08 +2026-08-25 22:41:44 | INFO | train rank=0 | step=50750/100000 progress=50.7% loss=0.126221 lr=5.289196e-06 elapsed=18:46:14 avg_step=1.33s recent_step=1.54s eta=18:12:57 +2026-08-25 22:42:00 | INFO | train rank=0 | step=50760/100000 progress=50.8% loss=0.044891 lr=5.287545e-06 elapsed=18:46:29 avg_step=1.33s recent_step=1.52s eta=18:12:46 +2026-08-25 22:42:17 | INFO | train rank=0 | step=50770/100000 progress=50.8% loss=0.047218 lr=5.285894e-06 elapsed=18:46:47 avg_step=1.33s recent_step=1.70s eta=18:12:36 +2026-08-25 22:42:31 | INFO | train rank=0 | step=50780/100000 progress=50.8% loss=0.042847 lr=5.284243e-06 elapsed=18:47:01 avg_step=1.33s recent_step=1.41s eta=18:12:23 +2026-08-25 22:42:42 | INFO | train rank=0 | step=50790/100000 progress=50.8% loss=0.047897 lr=5.282593e-06 elapsed=18:47:12 avg_step=1.33s recent_step=1.13s eta=18:12:08 +2026-08-25 22:42:53 | INFO | train rank=0 | step=50800/100000 progress=50.8% loss=0.092407 lr=5.280942e-06 elapsed=18:47:23 avg_step=1.33s recent_step=1.12s eta=18:11:53 +2026-08-25 22:43:08 | INFO | train rank=0 | step=50810/100000 progress=50.8% loss=0.075500 lr=5.279291e-06 elapsed=18:47:38 avg_step=1.33s recent_step=1.45s eta=18:11:40 +2026-08-25 22:43:20 | INFO | train rank=0 | step=50820/100000 progress=50.8% loss=0.060471 lr=5.277640e-06 elapsed=18:47:50 avg_step=1.33s recent_step=1.22s eta=18:11:26 +2026-08-25 22:43:32 | INFO | train rank=0 | step=50830/100000 progress=50.8% loss=0.062439 lr=5.275989e-06 elapsed=18:48:02 avg_step=1.33s recent_step=1.21s eta=18:11:12 +2026-08-25 22:43:48 | INFO | train rank=0 | step=50840/100000 progress=50.8% loss=0.099213 lr=5.274338e-06 elapsed=18:48:18 avg_step=1.33s recent_step=1.62s eta=18:11:01 +2026-08-25 22:44:03 | INFO | train rank=0 | step=50850/100000 progress=50.8% loss=0.048798 lr=5.272687e-06 elapsed=18:48:33 avg_step=1.33s recent_step=1.47s eta=18:10:49 +2026-08-25 22:44:20 | INFO | train rank=0 | step=50860/100000 progress=50.9% loss=0.036987 lr=5.271036e-06 elapsed=18:48:50 avg_step=1.33s recent_step=1.73s eta=18:10:40 +2026-08-25 22:44:33 | INFO | train rank=0 | step=50870/100000 progress=50.9% loss=0.045456 lr=5.269385e-06 elapsed=18:49:02 avg_step=1.33s recent_step=1.24s eta=18:10:25 +2026-08-25 22:44:44 | INFO | train rank=0 | step=50880/100000 progress=50.9% loss=0.045059 lr=5.267734e-06 elapsed=18:49:14 avg_step=1.33s recent_step=1.14s eta=18:10:10 +2026-08-25 22:45:03 | INFO | train rank=0 | step=50890/100000 progress=50.9% loss=0.060730 lr=5.266083e-06 elapsed=18:49:33 avg_step=1.33s recent_step=1.91s eta=18:10:02 +2026-08-25 22:45:14 | INFO | train rank=0 | step=50900/100000 progress=50.9% loss=0.044113 lr=5.264432e-06 elapsed=18:49:44 avg_step=1.33s recent_step=1.10s eta=18:09:47 +2026-08-25 22:45:27 | INFO | train rank=0 | step=50910/100000 progress=50.9% loss=0.055420 lr=5.262781e-06 elapsed=18:49:57 avg_step=1.33s recent_step=1.25s eta=18:09:33 +2026-08-25 22:45:38 | INFO | train rank=0 | step=50920/100000 progress=50.9% loss=0.033882 lr=5.261129e-06 elapsed=18:50:08 avg_step=1.33s recent_step=1.17s eta=18:09:18 +2026-08-25 22:45:50 | INFO | train rank=0 | step=50930/100000 progress=50.9% loss=0.062042 lr=5.259478e-06 elapsed=18:50:20 avg_step=1.33s recent_step=1.15s eta=18:09:03 +2026-08-25 22:46:04 | INFO | train rank=0 | step=50940/100000 progress=50.9% loss=0.031097 lr=5.257827e-06 elapsed=18:50:34 avg_step=1.33s recent_step=1.40s eta=18:08:50 +2026-08-25 22:46:15 | INFO | train rank=0 | step=50950/100000 progress=50.9% loss=0.040695 lr=5.256176e-06 elapsed=18:50:45 avg_step=1.33s recent_step=1.12s eta=18:08:35 +2026-08-25 22:46:31 | INFO | train rank=0 | step=50960/100000 progress=51.0% loss=0.034653 lr=5.254524e-06 elapsed=18:51:01 avg_step=1.33s recent_step=1.56s eta=18:08:24 +2026-08-25 22:46:42 | INFO | train rank=0 | step=50970/100000 progress=51.0% loss=0.111633 lr=5.252873e-06 elapsed=18:51:12 avg_step=1.33s recent_step=1.13s eta=18:08:09 +2026-08-25 22:46:53 | INFO | train rank=0 | step=50980/100000 progress=51.0% loss=0.035774 lr=5.251222e-06 elapsed=18:51:23 avg_step=1.33s recent_step=1.14s eta=18:07:53 +2026-08-25 22:47:07 | INFO | train rank=0 | step=50990/100000 progress=51.0% loss=0.039566 lr=5.249570e-06 elapsed=18:51:37 avg_step=1.33s recent_step=1.34s eta=18:07:40 +2026-08-25 22:47:21 | INFO | train rank=0 | step=51000/100000 progress=51.0% loss=0.035652 lr=5.247919e-06 elapsed=18:51:51 avg_step=1.33s recent_step=1.39s eta=18:07:27 +2026-08-25 22:47:32 | INFO | train rank=0 | step=51010/100000 progress=51.0% loss=0.060150 lr=5.246267e-06 elapsed=18:52:02 avg_step=1.33s recent_step=1.15s eta=18:07:12 +2026-08-25 22:47:48 | INFO | train rank=0 | step=51020/100000 progress=51.0% loss=0.060577 lr=5.244616e-06 elapsed=18:52:18 avg_step=1.33s recent_step=1.54s eta=18:07:01 +2026-08-25 22:48:02 | INFO | train rank=0 | step=51030/100000 progress=51.0% loss=0.047028 lr=5.242964e-06 elapsed=18:52:32 avg_step=1.33s recent_step=1.44s eta=18:06:49 +2026-08-25 22:48:17 | INFO | train rank=0 | step=51040/100000 progress=51.0% loss=0.038437 lr=5.241313e-06 elapsed=18:52:46 avg_step=1.33s recent_step=1.45s eta=18:06:37 +2026-08-25 22:48:28 | INFO | train rank=0 | step=51050/100000 progress=51.0% loss=0.049026 lr=5.239661e-06 elapsed=18:52:58 avg_step=1.33s recent_step=1.11s eta=18:06:21 +2026-08-25 22:48:39 | INFO | train rank=0 | step=51060/100000 progress=51.1% loss=0.062088 lr=5.238010e-06 elapsed=18:53:09 avg_step=1.33s recent_step=1.17s eta=18:06:06 +2026-08-25 22:48:54 | INFO | train rank=0 | step=51070/100000 progress=51.1% loss=0.052246 lr=5.236358e-06 elapsed=18:53:24 avg_step=1.33s recent_step=1.48s eta=18:05:54 +2026-08-25 22:49:08 | INFO | train rank=0 | step=51080/100000 progress=51.1% loss=0.036224 lr=5.234706e-06 elapsed=18:53:38 avg_step=1.33s recent_step=1.39s eta=18:05:42 +2026-08-25 22:49:19 | INFO | train rank=0 | step=51090/100000 progress=51.1% loss=0.073135 lr=5.233055e-06 elapsed=18:53:49 avg_step=1.33s recent_step=1.14s eta=18:05:27 +2026-08-25 22:49:33 | INFO | train rank=0 | step=51100/100000 progress=51.1% loss=0.059631 lr=5.231403e-06 elapsed=18:54:03 avg_step=1.33s recent_step=1.40s eta=18:05:14 +2026-08-25 22:49:46 | INFO | train rank=0 | step=51110/100000 progress=51.1% loss=0.038284 lr=5.229751e-06 elapsed=18:54:16 avg_step=1.33s recent_step=1.28s eta=18:05:00 +2026-08-25 22:49:59 | INFO | train rank=0 | step=51120/100000 progress=51.1% loss=0.055588 lr=5.228100e-06 elapsed=18:54:29 avg_step=1.33s recent_step=1.26s eta=18:04:46 +2026-08-25 22:50:13 | INFO | train rank=0 | step=51130/100000 progress=51.1% loss=0.055298 lr=5.226448e-06 elapsed=18:54:43 avg_step=1.33s recent_step=1.37s eta=18:04:33 +2026-08-25 22:50:24 | INFO | train rank=0 | step=51140/100000 progress=51.1% loss=0.041962 lr=5.224796e-06 elapsed=18:54:54 avg_step=1.33s recent_step=1.13s eta=18:04:18 +2026-08-25 22:50:40 | INFO | train rank=0 | step=51150/100000 progress=51.1% loss=0.034485 lr=5.223144e-06 elapsed=18:55:10 avg_step=1.33s recent_step=1.63s eta=18:04:08 +2026-08-25 22:50:56 | INFO | train rank=0 | step=51160/100000 progress=51.2% loss=0.057266 lr=5.221492e-06 elapsed=18:55:26 avg_step=1.33s recent_step=1.58s eta=18:03:57 +2026-08-25 22:51:08 | INFO | train rank=0 | step=51170/100000 progress=51.2% loss=0.047318 lr=5.219841e-06 elapsed=18:55:38 avg_step=1.33s recent_step=1.16s eta=18:03:42 +2026-08-25 22:51:20 | INFO | train rank=0 | step=51180/100000 progress=51.2% loss=0.047821 lr=5.218189e-06 elapsed=18:55:50 avg_step=1.33s recent_step=1.27s eta=18:03:28 +2026-08-25 22:51:35 | INFO | train rank=0 | step=51190/100000 progress=51.2% loss=0.060394 lr=5.216537e-06 elapsed=18:56:05 avg_step=1.33s recent_step=1.44s eta=18:03:15 +2026-08-25 22:51:50 | INFO | train rank=0 | step=51200/100000 progress=51.2% loss=0.027302 lr=5.214885e-06 elapsed=18:56:20 avg_step=1.33s recent_step=1.57s eta=18:03:04 +2026-08-25 22:52:06 | INFO | train rank=0 | step=51210/100000 progress=51.2% loss=0.052979 lr=5.213233e-06 elapsed=18:56:36 avg_step=1.33s recent_step=1.56s eta=18:02:53 +2026-08-25 22:52:21 | INFO | train rank=0 | step=51220/100000 progress=51.2% loss=0.056595 lr=5.211581e-06 elapsed=18:56:51 avg_step=1.33s recent_step=1.53s eta=18:02:42 +2026-08-25 22:52:36 | INFO | train rank=0 | step=51230/100000 progress=51.2% loss=0.029541 lr=5.209929e-06 elapsed=18:57:06 avg_step=1.33s recent_step=1.47s eta=18:02:30 +2026-08-25 22:52:47 | INFO | train rank=0 | step=51240/100000 progress=51.2% loss=0.087128 lr=5.208277e-06 elapsed=18:57:17 avg_step=1.33s recent_step=1.15s eta=18:02:15 +2026-08-25 22:52:59 | INFO | train rank=0 | step=51250/100000 progress=51.2% loss=0.058228 lr=5.206625e-06 elapsed=18:57:29 avg_step=1.33s recent_step=1.17s eta=18:02:00 +2026-08-25 22:53:12 | INFO | train rank=0 | step=51260/100000 progress=51.3% loss=0.035316 lr=5.204973e-06 elapsed=18:57:42 avg_step=1.33s recent_step=1.27s eta=18:01:46 +2026-08-25 22:53:28 | INFO | train rank=0 | step=51270/100000 progress=51.3% loss=0.044266 lr=5.203321e-06 elapsed=18:57:58 avg_step=1.33s recent_step=1.62s eta=18:01:35 +2026-08-25 22:53:44 | INFO | train rank=0 | step=51280/100000 progress=51.3% loss=0.046295 lr=5.201669e-06 elapsed=18:58:14 avg_step=1.33s recent_step=1.62s eta=18:01:25 +2026-08-25 22:53:59 | INFO | train rank=0 | step=51290/100000 progress=51.3% loss=0.046539 lr=5.200016e-06 elapsed=18:58:29 avg_step=1.33s recent_step=1.51s eta=18:01:13 +2026-08-25 22:54:11 | INFO | train rank=0 | step=51300/100000 progress=51.3% loss=0.066467 lr=5.198364e-06 elapsed=18:58:41 avg_step=1.33s recent_step=1.21s eta=18:00:59 +2026-08-25 22:54:25 | INFO | train rank=0 | step=51310/100000 progress=51.3% loss=0.046631 lr=5.196712e-06 elapsed=18:58:55 avg_step=1.33s recent_step=1.36s eta=18:00:46 +2026-08-25 22:54:40 | INFO | train rank=0 | step=51320/100000 progress=51.3% loss=0.037537 lr=5.195060e-06 elapsed=18:59:10 avg_step=1.33s recent_step=1.52s eta=18:00:34 +2026-08-25 22:54:54 | INFO | train rank=0 | step=51330/100000 progress=51.3% loss=0.063660 lr=5.193408e-06 elapsed=18:59:24 avg_step=1.33s recent_step=1.41s eta=18:00:21 +2026-08-25 22:55:05 | INFO | train rank=0 | step=51340/100000 progress=51.3% loss=0.054398 lr=5.191755e-06 elapsed=18:59:35 avg_step=1.33s recent_step=1.11s eta=18:00:06 +2026-08-25 22:55:17 | INFO | train rank=0 | step=51350/100000 progress=51.3% loss=0.049072 lr=5.190103e-06 elapsed=18:59:47 avg_step=1.33s recent_step=1.18s eta=17:59:51 +2026-08-25 22:55:28 | INFO | train rank=0 | step=51360/100000 progress=51.4% loss=0.055725 lr=5.188451e-06 elapsed=18:59:58 avg_step=1.33s recent_step=1.09s eta=17:59:36 +2026-08-25 22:55:40 | INFO | train rank=0 | step=51370/100000 progress=51.4% loss=0.042130 lr=5.186799e-06 elapsed=19:00:10 avg_step=1.33s recent_step=1.18s eta=17:59:21 +2026-08-25 22:55:54 | INFO | train rank=0 | step=51380/100000 progress=51.4% loss=0.048065 lr=5.185146e-06 elapsed=19:00:24 avg_step=1.33s recent_step=1.39s eta=17:59:08 +2026-08-25 22:56:05 | INFO | train rank=0 | step=51390/100000 progress=51.4% loss=0.036133 lr=5.183494e-06 elapsed=19:00:35 avg_step=1.33s recent_step=1.15s eta=17:58:53 +2026-08-25 22:56:20 | INFO | train rank=0 | step=51400/100000 progress=51.4% loss=0.061539 lr=5.181842e-06 elapsed=19:00:50 avg_step=1.33s recent_step=1.48s eta=17:58:41 +2026-08-25 22:56:36 | INFO | train rank=0 | step=51410/100000 progress=51.4% loss=0.093811 lr=5.180189e-06 elapsed=19:01:06 avg_step=1.33s recent_step=1.61s eta=17:58:30 +2026-08-25 22:56:47 | INFO | train rank=0 | step=51420/100000 progress=51.4% loss=0.042542 lr=5.178537e-06 elapsed=19:01:17 avg_step=1.33s recent_step=1.10s eta=17:58:15 +2026-08-25 22:57:01 | INFO | train rank=0 | step=51430/100000 progress=51.4% loss=0.062775 lr=5.176884e-06 elapsed=19:01:31 avg_step=1.33s recent_step=1.43s eta=17:58:03 +2026-08-25 22:57:13 | INFO | train rank=0 | step=51440/100000 progress=51.4% loss=0.065796 lr=5.175232e-06 elapsed=19:01:43 avg_step=1.33s recent_step=1.12s eta=17:57:47 +2026-08-25 22:57:25 | INFO | train rank=0 | step=51450/100000 progress=51.4% loss=0.060455 lr=5.173579e-06 elapsed=19:01:54 avg_step=1.33s recent_step=1.19s eta=17:57:33 +2026-08-25 22:57:38 | INFO | train rank=0 | step=51460/100000 progress=51.5% loss=0.073822 lr=5.171927e-06 elapsed=19:02:08 avg_step=1.33s recent_step=1.36s eta=17:57:20 +2026-08-25 22:57:49 | INFO | train rank=0 | step=51470/100000 progress=51.5% loss=0.047058 lr=5.170274e-06 elapsed=19:02:19 avg_step=1.33s recent_step=1.10s eta=17:57:04 +2026-08-25 22:58:03 | INFO | train rank=0 | step=51480/100000 progress=51.5% loss=0.033516 lr=5.168622e-06 elapsed=19:02:33 avg_step=1.33s recent_step=1.41s eta=17:56:51 +2026-08-25 22:58:19 | INFO | train rank=0 | step=51490/100000 progress=51.5% loss=0.055649 lr=5.166969e-06 elapsed=19:02:49 avg_step=1.33s recent_step=1.54s eta=17:56:40 +2026-08-25 22:58:35 | INFO | train rank=0 | step=51500/100000 progress=51.5% loss=0.075104 lr=5.165317e-06 elapsed=19:03:05 avg_step=1.33s recent_step=1.60s eta=17:56:29 +2026-08-25 22:58:52 | INFO | train rank=0 | step=51510/100000 progress=51.5% loss=0.037643 lr=5.163664e-06 elapsed=19:03:22 avg_step=1.33s recent_step=1.70s eta=17:56:19 +2026-08-25 22:59:03 | INFO | train rank=0 | step=51520/100000 progress=51.5% loss=0.039429 lr=5.162012e-06 elapsed=19:03:33 avg_step=1.33s recent_step=1.12s eta=17:56:04 +2026-08-25 22:59:15 | INFO | train rank=0 | step=51530/100000 progress=51.5% loss=0.044876 lr=5.160359e-06 elapsed=19:03:45 avg_step=1.33s recent_step=1.19s eta=17:55:49 +2026-08-25 22:59:26 | INFO | train rank=0 | step=51540/100000 progress=51.5% loss=0.023964 lr=5.158706e-06 elapsed=19:03:56 avg_step=1.33s recent_step=1.16s eta=17:55:35 +2026-08-25 22:59:40 | INFO | train rank=0 | step=51550/100000 progress=51.5% loss=0.044403 lr=5.157054e-06 elapsed=19:04:10 avg_step=1.33s recent_step=1.35s eta=17:55:21 +2026-08-25 22:59:58 | INFO | train rank=0 | step=51560/100000 progress=51.6% loss=0.052109 lr=5.155401e-06 elapsed=19:04:28 avg_step=1.33s recent_step=1.82s eta=17:55:13 +2026-08-25 23:00:09 | INFO | train rank=0 | step=51570/100000 progress=51.6% loss=0.073608 lr=5.153748e-06 elapsed=19:04:39 avg_step=1.33s recent_step=1.13s eta=17:54:58 +2026-08-25 23:00:21 | INFO | train rank=0 | step=51580/100000 progress=51.6% loss=0.053436 lr=5.152096e-06 elapsed=19:04:51 avg_step=1.33s recent_step=1.16s eta=17:54:43 +2026-08-25 23:00:33 | INFO | train rank=0 | step=51590/100000 progress=51.6% loss=0.042206 lr=5.150443e-06 elapsed=19:05:03 avg_step=1.33s recent_step=1.16s eta=17:54:28 +2026-08-25 23:00:47 | INFO | train rank=0 | step=51600/100000 progress=51.6% loss=0.086670 lr=5.148790e-06 elapsed=19:05:17 avg_step=1.33s recent_step=1.43s eta=17:54:15 +2026-08-25 23:01:02 | INFO | train rank=0 | step=51610/100000 progress=51.6% loss=0.040833 lr=5.147138e-06 elapsed=19:05:32 avg_step=1.33s recent_step=1.48s eta=17:54:03 +2026-08-25 23:01:16 | INFO | train rank=0 | step=51620/100000 progress=51.6% loss=0.035248 lr=5.145485e-06 elapsed=19:05:46 avg_step=1.33s recent_step=1.43s eta=17:53:51 +2026-08-25 23:01:31 | INFO | train rank=0 | step=51630/100000 progress=51.6% loss=0.049103 lr=5.143832e-06 elapsed=19:06:01 avg_step=1.33s recent_step=1.50s eta=17:53:39 +2026-08-25 23:01:43 | INFO | train rank=0 | step=51640/100000 progress=51.6% loss=0.042984 lr=5.142179e-06 elapsed=19:06:13 avg_step=1.33s recent_step=1.20s eta=17:53:25 +2026-08-25 23:01:54 | INFO | train rank=0 | step=51650/100000 progress=51.6% loss=0.062943 lr=5.140526e-06 elapsed=19:06:24 avg_step=1.33s recent_step=1.12s eta=17:53:09 +2026-08-25 23:02:09 | INFO | train rank=0 | step=51660/100000 progress=51.7% loss=0.037262 lr=5.138874e-06 elapsed=19:06:39 avg_step=1.33s recent_step=1.48s eta=17:52:57 +2026-08-25 23:02:24 | INFO | train rank=0 | step=51670/100000 progress=51.7% loss=0.063782 lr=5.137221e-06 elapsed=19:06:53 avg_step=1.33s recent_step=1.45s eta=17:52:45 +2026-08-25 23:02:38 | INFO | train rank=0 | step=51680/100000 progress=51.7% loss=0.068054 lr=5.135568e-06 elapsed=19:07:07 avg_step=1.33s recent_step=1.40s eta=17:52:33 +2026-08-25 23:02:53 | INFO | train rank=0 | step=51690/100000 progress=51.7% loss=0.055092 lr=5.133915e-06 elapsed=19:07:23 avg_step=1.33s recent_step=1.54s eta=17:52:21 +2026-08-25 23:03:08 | INFO | train rank=0 | step=51700/100000 progress=51.7% loss=0.117432 lr=5.132262e-06 elapsed=19:07:38 avg_step=1.33s recent_step=1.52s eta=17:52:10 +2026-08-25 23:03:20 | INFO | train rank=0 | step=51710/100000 progress=51.7% loss=0.038383 lr=5.130609e-06 elapsed=19:07:50 avg_step=1.33s recent_step=1.16s eta=17:51:55 +2026-08-25 23:03:34 | INFO | train rank=0 | step=51720/100000 progress=51.7% loss=0.064911 lr=5.128956e-06 elapsed=19:08:04 avg_step=1.33s recent_step=1.43s eta=17:51:42 +2026-08-25 23:03:51 | INFO | train rank=0 | step=51730/100000 progress=51.7% loss=0.070190 lr=5.127303e-06 elapsed=19:08:21 avg_step=1.33s recent_step=1.67s eta=17:51:32 +2026-08-25 23:04:04 | INFO | train rank=0 | step=51740/100000 progress=51.7% loss=0.036118 lr=5.125650e-06 elapsed=19:08:34 avg_step=1.33s recent_step=1.31s eta=17:51:19 +2026-08-25 23:04:15 | INFO | train rank=0 | step=51750/100000 progress=51.7% loss=0.058014 lr=5.123998e-06 elapsed=19:08:45 avg_step=1.33s recent_step=1.17s eta=17:51:04 +2026-08-25 23:04:27 | INFO | train rank=0 | step=51760/100000 progress=51.8% loss=0.031815 lr=5.122345e-06 elapsed=19:08:57 avg_step=1.33s recent_step=1.16s eta=17:50:49 +2026-08-25 23:04:39 | INFO | train rank=0 | step=51770/100000 progress=51.8% loss=0.094788 lr=5.120692e-06 elapsed=19:09:09 avg_step=1.33s recent_step=1.20s eta=17:50:34 +2026-08-25 23:04:52 | INFO | train rank=0 | step=51780/100000 progress=51.8% loss=0.046310 lr=5.119039e-06 elapsed=19:09:22 avg_step=1.33s recent_step=1.31s eta=17:50:21 +2026-08-25 23:05:10 | INFO | train rank=0 | step=51790/100000 progress=51.8% loss=0.057617 lr=5.117386e-06 elapsed=19:09:40 avg_step=1.33s recent_step=1.79s eta=17:50:12 +2026-08-25 23:05:22 | INFO | train rank=0 | step=51800/100000 progress=51.8% loss=0.043541 lr=5.115733e-06 elapsed=19:09:52 avg_step=1.33s recent_step=1.18s eta=17:49:57 +2026-08-25 23:05:34 | INFO | train rank=0 | step=51810/100000 progress=51.8% loss=0.040359 lr=5.114080e-06 elapsed=19:10:04 avg_step=1.33s recent_step=1.24s eta=17:49:43 +2026-08-25 23:05:48 | INFO | train rank=0 | step=51820/100000 progress=51.8% loss=0.053162 lr=5.112426e-06 elapsed=19:10:18 avg_step=1.33s recent_step=1.42s eta=17:49:30 +2026-08-25 23:06:01 | INFO | train rank=0 | step=51830/100000 progress=51.8% loss=0.037964 lr=5.110773e-06 elapsed=19:10:31 avg_step=1.33s recent_step=1.23s eta=17:49:16 +2026-08-25 23:06:19 | INFO | train rank=0 | step=51840/100000 progress=51.8% loss=0.047165 lr=5.109120e-06 elapsed=19:10:49 avg_step=1.33s recent_step=1.80s eta=17:49:07 +2026-08-25 23:06:31 | INFO | train rank=0 | step=51850/100000 progress=51.8% loss=0.068069 lr=5.107467e-06 elapsed=19:11:00 avg_step=1.33s recent_step=1.18s eta=17:48:52 +2026-08-25 23:06:42 | INFO | train rank=0 | step=51860/100000 progress=51.9% loss=0.070831 lr=5.105814e-06 elapsed=19:11:12 avg_step=1.33s recent_step=1.19s eta=17:48:38 +2026-08-25 23:06:54 | INFO | train rank=0 | step=51870/100000 progress=51.9% loss=0.035652 lr=5.104161e-06 elapsed=19:11:24 avg_step=1.33s recent_step=1.15s eta=17:48:23 +2026-08-25 23:07:10 | INFO | train rank=0 | step=51880/100000 progress=51.9% loss=0.056763 lr=5.102508e-06 elapsed=19:11:40 avg_step=1.33s recent_step=1.57s eta=17:48:11 +2026-08-25 23:07:25 | INFO | train rank=0 | step=51890/100000 progress=51.9% loss=0.050369 lr=5.100855e-06 elapsed=19:11:55 avg_step=1.33s recent_step=1.51s eta=17:48:00 +2026-08-25 23:07:37 | INFO | train rank=0 | step=51900/100000 progress=51.9% loss=0.073517 lr=5.099202e-06 elapsed=19:12:07 avg_step=1.33s recent_step=1.25s eta=17:47:46 +2026-08-25 23:07:50 | INFO | train rank=0 | step=51910/100000 progress=51.9% loss=0.048355 lr=5.097549e-06 elapsed=19:12:20 avg_step=1.33s recent_step=1.28s eta=17:47:32 +2026-08-25 23:08:06 | INFO | train rank=0 | step=51920/100000 progress=51.9% loss=0.052948 lr=5.095895e-06 elapsed=19:12:36 avg_step=1.33s recent_step=1.64s eta=17:47:21 +2026-08-25 23:08:19 | INFO | train rank=0 | step=51930/100000 progress=51.9% loss=0.046005 lr=5.094242e-06 elapsed=19:12:49 avg_step=1.33s recent_step=1.22s eta=17:47:07 +2026-08-25 23:08:30 | INFO | train rank=0 | step=51940/100000 progress=51.9% loss=0.041992 lr=5.092589e-06 elapsed=19:13:00 avg_step=1.33s recent_step=1.17s eta=17:46:52 +2026-08-25 23:08:42 | INFO | train rank=0 | step=51950/100000 progress=51.9% loss=0.050735 lr=5.090936e-06 elapsed=19:13:12 avg_step=1.33s recent_step=1.17s eta=17:46:37 +2026-08-25 23:08:54 | INFO | train rank=0 | step=51960/100000 progress=52.0% loss=0.038208 lr=5.089283e-06 elapsed=19:13:24 avg_step=1.33s recent_step=1.20s eta=17:46:23 +2026-08-25 23:09:05 | INFO | train rank=0 | step=51970/100000 progress=52.0% loss=0.041351 lr=5.087629e-06 elapsed=19:13:35 avg_step=1.33s recent_step=1.08s eta=17:46:07 +2026-08-25 23:09:19 | INFO | train rank=0 | step=51980/100000 progress=52.0% loss=0.056000 lr=5.085976e-06 elapsed=19:13:48 avg_step=1.33s recent_step=1.38s eta=17:45:54 +2026-08-25 23:09:34 | INFO | train rank=0 | step=51990/100000 progress=52.0% loss=0.030411 lr=5.084323e-06 elapsed=19:14:04 avg_step=1.33s recent_step=1.52s eta=17:45:43 +2026-08-25 23:09:46 | INFO | train rank=0 | step=52000/100000 progress=52.0% loss=0.048553 lr=5.082670e-06 elapsed=19:14:16 avg_step=1.33s recent_step=1.22s eta=17:45:28 +2026-08-25 23:10:01 | INFO | train rank=0 | step=52010/100000 progress=52.0% loss=0.042099 lr=5.081016e-06 elapsed=19:14:31 avg_step=1.33s recent_step=1.47s eta=17:45:16 +2026-08-25 23:10:13 | INFO | train rank=0 | step=52020/100000 progress=52.0% loss=0.057693 lr=5.079363e-06 elapsed=19:14:43 avg_step=1.33s recent_step=1.24s eta=17:45:02 +2026-08-25 23:10:25 | INFO | train rank=0 | step=52030/100000 progress=52.0% loss=0.061844 lr=5.077710e-06 elapsed=19:14:55 avg_step=1.33s recent_step=1.19s eta=17:44:48 +2026-08-25 23:10:40 | INFO | train rank=0 | step=52040/100000 progress=52.0% loss=0.067993 lr=5.076057e-06 elapsed=19:15:10 avg_step=1.33s recent_step=1.47s eta=17:44:36 +2026-08-25 23:10:51 | INFO | train rank=0 | step=52050/100000 progress=52.0% loss=0.026661 lr=5.074403e-06 elapsed=19:15:21 avg_step=1.33s recent_step=1.10s eta=17:44:20 +2026-08-25 23:11:05 | INFO | train rank=0 | step=52060/100000 progress=52.1% loss=0.066284 lr=5.072750e-06 elapsed=19:15:35 avg_step=1.33s recent_step=1.42s eta=17:44:08 +2026-08-25 23:11:19 | INFO | train rank=0 | step=52070/100000 progress=52.1% loss=0.048538 lr=5.071097e-06 elapsed=19:15:49 avg_step=1.33s recent_step=1.43s eta=17:43:55 +2026-08-25 23:11:33 | INFO | train rank=0 | step=52080/100000 progress=52.1% loss=0.056396 lr=5.069443e-06 elapsed=19:16:03 avg_step=1.33s recent_step=1.39s eta=17:43:42 +2026-08-25 23:11:45 | INFO | train rank=0 | step=52090/100000 progress=52.1% loss=0.026558 lr=5.067790e-06 elapsed=19:16:15 avg_step=1.33s recent_step=1.24s eta=17:43:28 +2026-08-25 23:12:00 | INFO | train rank=0 | step=52100/100000 progress=52.1% loss=0.066833 lr=5.066137e-06 elapsed=19:16:29 avg_step=1.33s recent_step=1.42s eta=17:43:16 +2026-08-25 23:12:16 | INFO | train rank=0 | step=52110/100000 progress=52.1% loss=0.025085 lr=5.064484e-06 elapsed=19:16:46 avg_step=1.33s recent_step=1.63s eta=17:43:05 +2026-08-25 23:12:28 | INFO | train rank=0 | step=52120/100000 progress=52.1% loss=0.053497 lr=5.062830e-06 elapsed=19:16:58 avg_step=1.33s recent_step=1.21s eta=17:42:51 +2026-08-25 23:12:41 | INFO | train rank=0 | step=52130/100000 progress=52.1% loss=0.040863 lr=5.061177e-06 elapsed=19:17:10 avg_step=1.33s recent_step=1.25s eta=17:42:37 +2026-08-25 23:12:52 | INFO | train rank=0 | step=52140/100000 progress=52.1% loss=0.042740 lr=5.059524e-06 elapsed=19:17:22 avg_step=1.33s recent_step=1.18s eta=17:42:22 +2026-08-25 23:13:06 | INFO | train rank=0 | step=52150/100000 progress=52.1% loss=0.111816 lr=5.057870e-06 elapsed=19:17:36 avg_step=1.33s recent_step=1.33s eta=17:42:09 +2026-08-25 23:13:16 | INFO | train rank=0 | step=52160/100000 progress=52.2% loss=0.052200 lr=5.056217e-06 elapsed=19:17:46 avg_step=1.33s recent_step=1.09s eta=17:41:53 +2026-08-25 23:13:31 | INFO | train rank=0 | step=52170/100000 progress=52.2% loss=0.028267 lr=5.054563e-06 elapsed=19:18:01 avg_step=1.33s recent_step=1.45s eta=17:41:41 +2026-08-25 23:13:44 | INFO | train rank=0 | step=52180/100000 progress=52.2% loss=0.058441 lr=5.052910e-06 elapsed=19:18:14 avg_step=1.33s recent_step=1.35s eta=17:41:28 +2026-08-25 23:13:56 | INFO | train rank=0 | step=52190/100000 progress=52.2% loss=0.051575 lr=5.051257e-06 elapsed=19:18:26 avg_step=1.33s recent_step=1.17s eta=17:41:13 +2026-08-25 23:14:11 | INFO | train rank=0 | step=52200/100000 progress=52.2% loss=0.056213 lr=5.049603e-06 elapsed=19:18:41 avg_step=1.33s recent_step=1.52s eta=17:41:01 +2026-08-25 23:14:22 | INFO | train rank=0 | step=52210/100000 progress=52.2% loss=0.054077 lr=5.047950e-06 elapsed=19:18:52 avg_step=1.33s recent_step=1.10s eta=17:40:46 +2026-08-25 23:14:40 | INFO | train rank=0 | step=52220/100000 progress=52.2% loss=0.038391 lr=5.046296e-06 elapsed=19:19:10 avg_step=1.33s recent_step=1.81s eta=17:40:37 +2026-08-25 23:14:52 | INFO | train rank=0 | step=52230/100000 progress=52.2% loss=0.052414 lr=5.044643e-06 elapsed=19:19:22 avg_step=1.33s recent_step=1.13s eta=17:40:22 +2026-08-25 23:15:04 | INFO | train rank=0 | step=52240/100000 progress=52.2% loss=0.039322 lr=5.042990e-06 elapsed=19:19:34 avg_step=1.33s recent_step=1.18s eta=17:40:07 +2026-08-25 23:15:20 | INFO | train rank=0 | step=52250/100000 progress=52.2% loss=0.047455 lr=5.041336e-06 elapsed=19:19:50 avg_step=1.33s recent_step=1.62s eta=17:39:56 +2026-08-25 23:15:34 | INFO | train rank=0 | step=52260/100000 progress=52.3% loss=0.078339 lr=5.039683e-06 elapsed=19:20:04 avg_step=1.33s recent_step=1.42s eta=17:39:44 +2026-08-25 23:15:49 | INFO | train rank=0 | step=52270/100000 progress=52.3% loss=0.041199 lr=5.038029e-06 elapsed=19:20:18 avg_step=1.33s recent_step=1.45s eta=17:39:32 +2026-08-25 23:16:03 | INFO | train rank=0 | step=52280/100000 progress=52.3% loss=0.044464 lr=5.036376e-06 elapsed=19:20:32 avg_step=1.33s recent_step=1.41s eta=17:39:19 +2026-08-25 23:16:14 | INFO | train rank=0 | step=52290/100000 progress=52.3% loss=0.072235 lr=5.034723e-06 elapsed=19:20:44 avg_step=1.33s recent_step=1.12s eta=17:39:04 +2026-08-25 23:16:30 | INFO | train rank=0 | step=52300/100000 progress=52.3% loss=0.060211 lr=5.033069e-06 elapsed=19:21:00 avg_step=1.33s recent_step=1.64s eta=17:38:53 +2026-08-25 23:16:41 | INFO | train rank=0 | step=52310/100000 progress=52.3% loss=0.032913 lr=5.031416e-06 elapsed=19:21:11 avg_step=1.33s recent_step=1.11s eta=17:38:38 +2026-08-25 23:16:53 | INFO | train rank=0 | step=52320/100000 progress=52.3% loss=0.065216 lr=5.029762e-06 elapsed=19:21:23 avg_step=1.33s recent_step=1.16s eta=17:38:23 +2026-08-25 23:17:04 | INFO | train rank=0 | step=52330/100000 progress=52.3% loss=0.050018 lr=5.028109e-06 elapsed=19:21:34 avg_step=1.33s recent_step=1.14s eta=17:38:08 +2026-08-25 23:17:17 | INFO | train rank=0 | step=52340/100000 progress=52.3% loss=0.054520 lr=5.026455e-06 elapsed=19:21:47 avg_step=1.33s recent_step=1.31s eta=17:37:54 +2026-08-25 23:17:30 | INFO | train rank=0 | step=52350/100000 progress=52.3% loss=0.045868 lr=5.024802e-06 elapsed=19:22:00 avg_step=1.33s recent_step=1.28s eta=17:37:41 +2026-08-25 23:17:47 | INFO | train rank=0 | step=52360/100000 progress=52.4% loss=0.068176 lr=5.023148e-06 elapsed=19:22:17 avg_step=1.33s recent_step=1.64s eta=17:37:30 +2026-08-25 23:17:58 | INFO | train rank=0 | step=52370/100000 progress=52.4% loss=0.050369 lr=5.021495e-06 elapsed=19:22:28 avg_step=1.33s recent_step=1.14s eta=17:37:15 +2026-08-25 23:18:13 | INFO | train rank=0 | step=52380/100000 progress=52.4% loss=0.037903 lr=5.019842e-06 elapsed=19:22:43 avg_step=1.33s recent_step=1.45s eta=17:37:03 +2026-08-25 23:18:24 | INFO | train rank=0 | step=52390/100000 progress=52.4% loss=0.062286 lr=5.018188e-06 elapsed=19:22:54 avg_step=1.33s recent_step=1.18s eta=17:36:48 +2026-08-25 23:18:40 | INFO | train rank=0 | step=52400/100000 progress=52.4% loss=0.056854 lr=5.016535e-06 elapsed=19:23:09 avg_step=1.33s recent_step=1.51s eta=17:36:37 +2026-08-25 23:18:59 | INFO | train rank=0 | step=52410/100000 progress=52.4% loss=0.054749 lr=5.014881e-06 elapsed=19:23:29 avg_step=1.33s recent_step=1.93s eta=17:36:29 +2026-08-25 23:19:10 | INFO | train rank=0 | step=52420/100000 progress=52.4% loss=0.032440 lr=5.013228e-06 elapsed=19:23:40 avg_step=1.33s recent_step=1.09s eta=17:36:13 +2026-08-25 23:19:26 | INFO | train rank=0 | step=52430/100000 progress=52.4% loss=0.045013 lr=5.011574e-06 elapsed=19:23:56 avg_step=1.33s recent_step=1.63s eta=17:36:02 +2026-08-25 23:19:41 | INFO | train rank=0 | step=52440/100000 progress=52.4% loss=0.055084 lr=5.009921e-06 elapsed=19:24:11 avg_step=1.33s recent_step=1.50s eta=17:35:51 +2026-08-25 23:19:55 | INFO | train rank=0 | step=52450/100000 progress=52.4% loss=0.074280 lr=5.008267e-06 elapsed=19:24:25 avg_step=1.33s recent_step=1.36s eta=17:35:38 +2026-08-25 23:20:08 | INFO | train rank=0 | step=52460/100000 progress=52.5% loss=0.045380 lr=5.006614e-06 elapsed=19:24:38 avg_step=1.33s recent_step=1.31s eta=17:35:24 +2026-08-25 23:20:23 | INFO | train rank=0 | step=52470/100000 progress=52.5% loss=0.057800 lr=5.004960e-06 elapsed=19:24:53 avg_step=1.33s recent_step=1.55s eta=17:35:13 +2026-08-25 23:20:35 | INFO | train rank=0 | step=52480/100000 progress=52.5% loss=0.045715 lr=5.003307e-06 elapsed=19:25:04 avg_step=1.33s recent_step=1.12s eta=17:34:58 +2026-08-25 23:20:47 | INFO | train rank=0 | step=52490/100000 progress=52.5% loss=0.061630 lr=5.001653e-06 elapsed=19:25:17 avg_step=1.33s recent_step=1.25s eta=17:34:43 +2026-08-25 23:21:05 | INFO | train rank=0 | step=52500/100000 progress=52.5% loss=0.060730 lr=5.000000e-06 elapsed=19:25:35 avg_step=1.33s recent_step=1.78s eta=17:34:34 +2026-08-25 23:21:17 | INFO | train rank=0 | step=52510/100000 progress=52.5% loss=0.055450 lr=4.998347e-06 elapsed=19:25:47 avg_step=1.33s recent_step=1.19s eta=17:34:20 +2026-08-25 23:21:28 | INFO | train rank=0 | step=52520/100000 progress=52.5% loss=0.036240 lr=4.996693e-06 elapsed=19:25:58 avg_step=1.33s recent_step=1.12s eta=17:34:04 +2026-08-25 23:21:39 | INFO | train rank=0 | step=52530/100000 progress=52.5% loss=0.097809 lr=4.995040e-06 elapsed=19:26:09 avg_step=1.33s recent_step=1.11s eta=17:33:49 +2026-08-25 23:21:53 | INFO | train rank=0 | step=52540/100000 progress=52.5% loss=0.047928 lr=4.993386e-06 elapsed=19:26:23 avg_step=1.33s recent_step=1.38s eta=17:33:36 +2026-08-25 23:22:04 | INFO | train rank=0 | step=52550/100000 progress=52.5% loss=0.041946 lr=4.991733e-06 elapsed=19:26:34 avg_step=1.33s recent_step=1.11s eta=17:33:21 +2026-08-25 23:22:16 | INFO | train rank=0 | step=52560/100000 progress=52.6% loss=0.047104 lr=4.990079e-06 elapsed=19:26:46 avg_step=1.33s recent_step=1.18s eta=17:33:06 +2026-08-25 23:22:30 | INFO | train rank=0 | step=52570/100000 progress=52.6% loss=0.044037 lr=4.988426e-06 elapsed=19:27:00 avg_step=1.33s recent_step=1.44s eta=17:32:54 +2026-08-25 23:22:44 | INFO | train rank=0 | step=52580/100000 progress=52.6% loss=0.067459 lr=4.986772e-06 elapsed=19:27:14 avg_step=1.33s recent_step=1.34s eta=17:32:41 +2026-08-25 23:22:55 | INFO | train rank=0 | step=52590/100000 progress=52.6% loss=0.052559 lr=4.985119e-06 elapsed=19:27:25 avg_step=1.33s recent_step=1.17s eta=17:32:26 +2026-08-25 23:23:09 | INFO | train rank=0 | step=52600/100000 progress=52.6% loss=0.080200 lr=4.983465e-06 elapsed=19:27:39 avg_step=1.33s recent_step=1.37s eta=17:32:13 +2026-08-25 23:23:23 | INFO | train rank=0 | step=52610/100000 progress=52.6% loss=0.046021 lr=4.981812e-06 elapsed=19:27:52 avg_step=1.33s recent_step=1.35s eta=17:32:00 +2026-08-25 23:23:40 | INFO | train rank=0 | step=52620/100000 progress=52.6% loss=0.060974 lr=4.980158e-06 elapsed=19:28:10 avg_step=1.33s recent_step=1.77s eta=17:31:50 +2026-08-25 23:23:57 | INFO | train rank=0 | step=52630/100000 progress=52.6% loss=0.064789 lr=4.978505e-06 elapsed=19:28:27 avg_step=1.33s recent_step=1.65s eta=17:31:40 +2026-08-25 23:24:09 | INFO | train rank=0 | step=52640/100000 progress=52.6% loss=0.067535 lr=4.976852e-06 elapsed=19:28:38 avg_step=1.33s recent_step=1.18s eta=17:31:25 +2026-08-25 23:24:20 | INFO | train rank=0 | step=52650/100000 progress=52.6% loss=0.040771 lr=4.975198e-06 elapsed=19:28:50 avg_step=1.33s recent_step=1.19s eta=17:31:11 +2026-08-25 23:24:35 | INFO | train rank=0 | step=52660/100000 progress=52.7% loss=0.067017 lr=4.973545e-06 elapsed=19:29:05 avg_step=1.33s recent_step=1.50s eta=17:30:59 +2026-08-25 23:24:53 | INFO | train rank=0 | step=52670/100000 progress=52.7% loss=0.035561 lr=4.971891e-06 elapsed=19:29:23 avg_step=1.33s recent_step=1.78s eta=17:30:50 +2026-08-25 23:25:05 | INFO | train rank=0 | step=52680/100000 progress=52.7% loss=0.102936 lr=4.970238e-06 elapsed=19:29:35 avg_step=1.33s recent_step=1.15s eta=17:30:35 +2026-08-25 23:25:16 | INFO | train rank=0 | step=52690/100000 progress=52.7% loss=0.074921 lr=4.968584e-06 elapsed=19:29:46 avg_step=1.33s recent_step=1.10s eta=17:30:19 +2026-08-25 23:25:32 | INFO | train rank=0 | step=52700/100000 progress=52.7% loss=0.047882 lr=4.966931e-06 elapsed=19:30:02 avg_step=1.33s recent_step=1.64s eta=17:30:09 +2026-08-25 23:25:47 | INFO | train rank=0 | step=52710/100000 progress=52.7% loss=0.062286 lr=4.965277e-06 elapsed=19:30:17 avg_step=1.33s recent_step=1.53s eta=17:29:57 +2026-08-25 23:26:02 | INFO | train rank=0 | step=52720/100000 progress=52.7% loss=0.063263 lr=4.963624e-06 elapsed=19:30:32 avg_step=1.33s recent_step=1.43s eta=17:29:45 +2026-08-25 23:26:15 | INFO | train rank=0 | step=52730/100000 progress=52.7% loss=0.046097 lr=4.961971e-06 elapsed=19:30:45 avg_step=1.33s recent_step=1.38s eta=17:29:32 +2026-08-25 23:26:27 | INFO | train rank=0 | step=52740/100000 progress=52.7% loss=0.044861 lr=4.960317e-06 elapsed=19:30:57 avg_step=1.33s recent_step=1.17s eta=17:29:17 +2026-08-25 23:26:41 | INFO | train rank=0 | step=52750/100000 progress=52.8% loss=0.048767 lr=4.958664e-06 elapsed=19:31:11 avg_step=1.33s recent_step=1.39s eta=17:29:04 +2026-08-25 23:26:54 | INFO | train rank=0 | step=52760/100000 progress=52.8% loss=0.030563 lr=4.957010e-06 elapsed=19:31:24 avg_step=1.33s recent_step=1.29s eta=17:28:50 +2026-08-25 23:27:05 | INFO | train rank=0 | step=52770/100000 progress=52.8% loss=0.044876 lr=4.955357e-06 elapsed=19:31:35 avg_step=1.33s recent_step=1.14s eta=17:28:35 +2026-08-25 23:27:23 | INFO | train rank=0 | step=52780/100000 progress=52.8% loss=0.073517 lr=4.953704e-06 elapsed=19:31:53 avg_step=1.33s recent_step=1.79s eta=17:28:26 +2026-08-25 23:27:35 | INFO | train rank=0 | step=52790/100000 progress=52.8% loss=0.037079 lr=4.952050e-06 elapsed=19:32:05 avg_step=1.33s recent_step=1.20s eta=17:28:12 +2026-08-25 23:27:47 | INFO | train rank=0 | step=52800/100000 progress=52.8% loss=0.059906 lr=4.950397e-06 elapsed=19:32:17 avg_step=1.33s recent_step=1.15s eta=17:27:57 +2026-08-25 23:28:03 | INFO | train rank=0 | step=52810/100000 progress=52.8% loss=0.079437 lr=4.948743e-06 elapsed=19:32:33 avg_step=1.33s recent_step=1.63s eta=17:27:46 +2026-08-25 23:28:15 | INFO | train rank=0 | step=52820/100000 progress=52.8% loss=0.028954 lr=4.947090e-06 elapsed=19:32:45 avg_step=1.33s recent_step=1.23s eta=17:27:32 +2026-08-25 23:28:30 | INFO | train rank=0 | step=52830/100000 progress=52.8% loss=0.057587 lr=4.945437e-06 elapsed=19:33:00 avg_step=1.33s recent_step=1.43s eta=17:27:19 +2026-08-25 23:28:45 | INFO | train rank=0 | step=52840/100000 progress=52.8% loss=0.031013 lr=4.943783e-06 elapsed=19:33:15 avg_step=1.33s recent_step=1.57s eta=17:27:08 +2026-08-25 23:28:57 | INFO | train rank=0 | step=52850/100000 progress=52.8% loss=0.041840 lr=4.942130e-06 elapsed=19:33:27 avg_step=1.33s recent_step=1.14s eta=17:26:53 +2026-08-25 23:29:09 | INFO | train rank=0 | step=52860/100000 progress=52.9% loss=0.036926 lr=4.940476e-06 elapsed=19:33:39 avg_step=1.33s recent_step=1.21s eta=17:26:39 +2026-08-25 23:29:22 | INFO | train rank=0 | step=52870/100000 progress=52.9% loss=0.046753 lr=4.938823e-06 elapsed=19:33:52 avg_step=1.33s recent_step=1.35s eta=17:26:26 +2026-08-25 23:29:34 | INFO | train rank=0 | step=52880/100000 progress=52.9% loss=0.060150 lr=4.937170e-06 elapsed=19:34:04 avg_step=1.33s recent_step=1.15s eta=17:26:11 +2026-08-25 23:29:46 | INFO | train rank=0 | step=52890/100000 progress=52.9% loss=0.052521 lr=4.935516e-06 elapsed=19:34:15 avg_step=1.33s recent_step=1.16s eta=17:25:56 +2026-08-25 23:29:58 | INFO | train rank=0 | step=52900/100000 progress=52.9% loss=0.048309 lr=4.933863e-06 elapsed=19:34:28 avg_step=1.33s recent_step=1.24s eta=17:25:42 +2026-08-25 23:30:13 | INFO | train rank=0 | step=52910/100000 progress=52.9% loss=0.033768 lr=4.932210e-06 elapsed=19:34:43 avg_step=1.33s recent_step=1.55s eta=17:25:30 +2026-08-25 23:30:29 | INFO | train rank=0 | step=52920/100000 progress=52.9% loss=0.048813 lr=4.930557e-06 elapsed=19:34:59 avg_step=1.33s recent_step=1.55s eta=17:25:19 +2026-08-25 23:30:44 | INFO | train rank=0 | step=52930/100000 progress=52.9% loss=0.040192 lr=4.928903e-06 elapsed=19:35:14 avg_step=1.33s recent_step=1.49s eta=17:25:07 +2026-08-25 23:30:56 | INFO | train rank=0 | step=52940/100000 progress=52.9% loss=0.072998 lr=4.927250e-06 elapsed=19:35:26 avg_step=1.33s recent_step=1.20s eta=17:24:52 +2026-08-25 23:31:10 | INFO | train rank=0 | step=52950/100000 progress=52.9% loss=0.067963 lr=4.925597e-06 elapsed=19:35:40 avg_step=1.33s recent_step=1.44s eta=17:24:40 +2026-08-25 23:31:22 | INFO | train rank=0 | step=52960/100000 progress=53.0% loss=0.035881 lr=4.923943e-06 elapsed=19:35:52 avg_step=1.33s recent_step=1.17s eta=17:24:25 +2026-08-25 23:31:36 | INFO | train rank=0 | step=52970/100000 progress=53.0% loss=0.053406 lr=4.922290e-06 elapsed=19:36:06 avg_step=1.33s recent_step=1.42s eta=17:24:13 +2026-08-25 23:31:51 | INFO | train rank=0 | step=52980/100000 progress=53.0% loss=0.046814 lr=4.920637e-06 elapsed=19:36:20 avg_step=1.33s recent_step=1.44s eta=17:24:00 +2026-08-25 23:32:02 | INFO | train rank=0 | step=52990/100000 progress=53.0% loss=0.032990 lr=4.918984e-06 elapsed=19:36:32 avg_step=1.33s recent_step=1.16s eta=17:23:46 +2026-08-25 23:32:17 | INFO | train rank=0 | step=53000/100000 progress=53.0% loss=0.061310 lr=4.917330e-06 elapsed=19:36:47 avg_step=1.33s recent_step=1.50s eta=17:23:34 +2026-08-25 23:32:29 | INFO | train rank=0 | step=53010/100000 progress=53.0% loss=0.059586 lr=4.915677e-06 elapsed=19:36:59 avg_step=1.33s recent_step=1.16s eta=17:23:19 +2026-08-25 23:32:44 | INFO | train rank=0 | step=53020/100000 progress=53.0% loss=0.066956 lr=4.914024e-06 elapsed=19:37:13 avg_step=1.33s recent_step=1.47s eta=17:23:07 +2026-08-25 23:32:57 | INFO | train rank=0 | step=53030/100000 progress=53.0% loss=0.115112 lr=4.912371e-06 elapsed=19:37:27 avg_step=1.33s recent_step=1.35s eta=17:22:54 +2026-08-25 23:33:09 | INFO | train rank=0 | step=53040/100000 progress=53.0% loss=0.067032 lr=4.910717e-06 elapsed=19:37:39 avg_step=1.33s recent_step=1.19s eta=17:22:39 +2026-08-25 23:33:25 | INFO | train rank=0 | step=53050/100000 progress=53.0% loss=0.032501 lr=4.909064e-06 elapsed=19:37:55 avg_step=1.33s recent_step=1.59s eta=17:22:28 +2026-08-25 23:33:36 | INFO | train rank=0 | step=53060/100000 progress=53.1% loss=0.037613 lr=4.907411e-06 elapsed=19:38:06 avg_step=1.33s recent_step=1.15s eta=17:22:13 +2026-08-25 23:33:49 | INFO | train rank=0 | step=53070/100000 progress=53.1% loss=0.041710 lr=4.905758e-06 elapsed=19:38:19 avg_step=1.33s recent_step=1.28s eta=17:21:59 +2026-08-25 23:34:01 | INFO | train rank=0 | step=53080/100000 progress=53.1% loss=0.067047 lr=4.904105e-06 elapsed=19:38:31 avg_step=1.33s recent_step=1.18s eta=17:21:45 +2026-08-25 23:34:18 | INFO | train rank=0 | step=53090/100000 progress=53.1% loss=0.050613 lr=4.902451e-06 elapsed=19:38:48 avg_step=1.33s recent_step=1.76s eta=17:21:35 +2026-08-25 23:34:30 | INFO | train rank=0 | step=53100/100000 progress=53.1% loss=0.036842 lr=4.900798e-06 elapsed=19:39:00 avg_step=1.33s recent_step=1.20s eta=17:21:20 +2026-08-25 23:34:45 | INFO | train rank=0 | step=53110/100000 progress=53.1% loss=0.038162 lr=4.899145e-06 elapsed=19:39:15 avg_step=1.33s recent_step=1.48s eta=17:21:08 +2026-08-25 23:34:57 | INFO | train rank=0 | step=53120/100000 progress=53.1% loss=0.076080 lr=4.897492e-06 elapsed=19:39:27 avg_step=1.33s recent_step=1.15s eta=17:20:54 +2026-08-25 23:35:10 | INFO | train rank=0 | step=53130/100000 progress=53.1% loss=0.050385 lr=4.895839e-06 elapsed=19:39:39 avg_step=1.33s recent_step=1.29s eta=17:20:40 +2026-08-25 23:35:21 | INFO | train rank=0 | step=53140/100000 progress=53.1% loss=0.074768 lr=4.894186e-06 elapsed=19:39:51 avg_step=1.33s recent_step=1.11s eta=17:20:25 +2026-08-25 23:35:32 | INFO | train rank=0 | step=53150/100000 progress=53.1% loss=0.038986 lr=4.892533e-06 elapsed=19:40:02 avg_step=1.33s recent_step=1.16s eta=17:20:10 +2026-08-25 23:35:50 | INFO | train rank=0 | step=53160/100000 progress=53.2% loss=0.061356 lr=4.890880e-06 elapsed=19:40:19 avg_step=1.33s recent_step=1.72s eta=17:20:00 +2026-08-25 23:36:01 | INFO | train rank=0 | step=53170/100000 progress=53.2% loss=0.045670 lr=4.889227e-06 elapsed=19:40:31 avg_step=1.33s recent_step=1.15s eta=17:19:45 +2026-08-25 23:36:21 | INFO | train rank=0 | step=53180/100000 progress=53.2% loss=0.045822 lr=4.887574e-06 elapsed=19:40:51 avg_step=1.33s recent_step=2.03s eta=17:19:38 +2026-08-25 23:36:33 | INFO | train rank=0 | step=53190/100000 progress=53.2% loss=0.074005 lr=4.885920e-06 elapsed=19:41:03 avg_step=1.33s recent_step=1.17s eta=17:19:23 +2026-08-25 23:36:49 | INFO | train rank=0 | step=53200/100000 progress=53.2% loss=0.087341 lr=4.884267e-06 elapsed=19:41:19 avg_step=1.33s recent_step=1.65s eta=17:19:12 +2026-08-25 23:37:01 | INFO | train rank=0 | step=53210/100000 progress=53.2% loss=0.044708 lr=4.882614e-06 elapsed=19:41:31 avg_step=1.33s recent_step=1.14s eta=17:18:57 +2026-08-25 23:37:13 | INFO | train rank=0 | step=53220/100000 progress=53.2% loss=0.038063 lr=4.880961e-06 elapsed=19:41:43 avg_step=1.33s recent_step=1.20s eta=17:18:43 +2026-08-25 23:37:29 | INFO | train rank=0 | step=53230/100000 progress=53.2% loss=0.040741 lr=4.879308e-06 elapsed=19:41:59 avg_step=1.33s recent_step=1.58s eta=17:18:32 +2026-08-25 23:37:44 | INFO | train rank=0 | step=53240/100000 progress=53.2% loss=0.043259 lr=4.877655e-06 elapsed=19:42:14 avg_step=1.33s recent_step=1.56s eta=17:18:20 +2026-08-25 23:37:56 | INFO | train rank=0 | step=53250/100000 progress=53.2% loss=0.051025 lr=4.876002e-06 elapsed=19:42:26 avg_step=1.33s recent_step=1.15s eta=17:18:05 +2026-08-25 23:38:07 | INFO | train rank=0 | step=53260/100000 progress=53.3% loss=0.040543 lr=4.874350e-06 elapsed=19:42:37 avg_step=1.33s recent_step=1.17s eta=17:17:51 +2026-08-25 23:38:19 | INFO | train rank=0 | step=53270/100000 progress=53.3% loss=0.034607 lr=4.872697e-06 elapsed=19:42:49 avg_step=1.33s recent_step=1.18s eta=17:17:36 +2026-08-25 23:38:34 | INFO | train rank=0 | step=53280/100000 progress=53.3% loss=0.076782 lr=4.871044e-06 elapsed=19:43:04 avg_step=1.33s recent_step=1.51s eta=17:17:24 +2026-08-25 23:38:46 | INFO | train rank=0 | step=53290/100000 progress=53.3% loss=0.048203 lr=4.869391e-06 elapsed=19:43:15 avg_step=1.33s recent_step=1.13s eta=17:17:09 +2026-08-25 23:39:00 | INFO | train rank=0 | step=53300/100000 progress=53.3% loss=0.056702 lr=4.867738e-06 elapsed=19:43:30 avg_step=1.33s recent_step=1.40s eta=17:16:57 +2026-08-25 23:39:12 | INFO | train rank=0 | step=53310/100000 progress=53.3% loss=0.040817 lr=4.866085e-06 elapsed=19:43:42 avg_step=1.33s recent_step=1.24s eta=17:16:42 +2026-08-25 23:39:25 | INFO | train rank=0 | step=53320/100000 progress=53.3% loss=0.062439 lr=4.864432e-06 elapsed=19:43:55 avg_step=1.33s recent_step=1.31s eta=17:16:29 +2026-08-25 23:39:41 | INFO | train rank=0 | step=53330/100000 progress=53.3% loss=0.062439 lr=4.862779e-06 elapsed=19:44:11 avg_step=1.33s recent_step=1.59s eta=17:16:18 +2026-08-25 23:39:53 | INFO | train rank=0 | step=53340/100000 progress=53.3% loss=0.046310 lr=4.861126e-06 elapsed=19:44:23 avg_step=1.33s recent_step=1.17s eta=17:16:03 +2026-08-25 23:40:07 | INFO | train rank=0 | step=53350/100000 progress=53.3% loss=0.049469 lr=4.859474e-06 elapsed=19:44:37 avg_step=1.33s recent_step=1.40s eta=17:15:50 +2026-08-25 23:40:22 | INFO | train rank=0 | step=53360/100000 progress=53.4% loss=0.027832 lr=4.857821e-06 elapsed=19:44:52 avg_step=1.33s recent_step=1.55s eta=17:15:39 +2026-08-25 23:40:35 | INFO | train rank=0 | step=53370/100000 progress=53.4% loss=0.033615 lr=4.856168e-06 elapsed=19:45:05 avg_step=1.33s recent_step=1.29s eta=17:15:25 +2026-08-25 23:40:49 | INFO | train rank=0 | step=53380/100000 progress=53.4% loss=0.035034 lr=4.854515e-06 elapsed=19:45:18 avg_step=1.33s recent_step=1.35s eta=17:15:12 +2026-08-25 23:41:03 | INFO | train rank=0 | step=53390/100000 progress=53.4% loss=0.043945 lr=4.852862e-06 elapsed=19:45:33 avg_step=1.33s recent_step=1.45s eta=17:15:00 +2026-08-25 23:41:16 | INFO | train rank=0 | step=53400/100000 progress=53.4% loss=0.039963 lr=4.851210e-06 elapsed=19:45:46 avg_step=1.33s recent_step=1.33s eta=17:14:46 +2026-08-25 23:41:28 | INFO | train rank=0 | step=53410/100000 progress=53.4% loss=0.049294 lr=4.849557e-06 elapsed=19:45:58 avg_step=1.33s recent_step=1.19s eta=17:14:32 +2026-08-25 23:41:40 | INFO | train rank=0 | step=53420/100000 progress=53.4% loss=0.068787 lr=4.847904e-06 elapsed=19:46:10 avg_step=1.33s recent_step=1.16s eta=17:14:17 +2026-08-25 23:41:54 | INFO | train rank=0 | step=53430/100000 progress=53.4% loss=0.075684 lr=4.846252e-06 elapsed=19:46:24 avg_step=1.33s recent_step=1.39s eta=17:14:04 +2026-08-25 23:42:05 | INFO | train rank=0 | step=53440/100000 progress=53.4% loss=0.048065 lr=4.844599e-06 elapsed=19:46:35 avg_step=1.33s recent_step=1.16s eta=17:13:49 +2026-08-25 23:42:17 | INFO | train rank=0 | step=53450/100000 progress=53.4% loss=0.074890 lr=4.842946e-06 elapsed=19:46:47 avg_step=1.33s recent_step=1.17s eta=17:13:35 +2026-08-25 23:42:29 | INFO | train rank=0 | step=53460/100000 progress=53.5% loss=0.048218 lr=4.841294e-06 elapsed=19:46:59 avg_step=1.33s recent_step=1.15s eta=17:13:20 +2026-08-25 23:42:40 | INFO | train rank=0 | step=53470/100000 progress=53.5% loss=0.051849 lr=4.839641e-06 elapsed=19:47:10 avg_step=1.33s recent_step=1.17s eta=17:13:05 +2026-08-25 23:42:56 | INFO | train rank=0 | step=53480/100000 progress=53.5% loss=0.054459 lr=4.837988e-06 elapsed=19:47:26 avg_step=1.33s recent_step=1.60s eta=17:12:54 +2026-08-25 23:43:11 | INFO | train rank=0 | step=53490/100000 progress=53.5% loss=0.062866 lr=4.836336e-06 elapsed=19:47:41 avg_step=1.33s recent_step=1.48s eta=17:12:42 +2026-08-25 23:43:25 | INFO | train rank=0 | step=53500/100000 progress=53.5% loss=0.047340 lr=4.834683e-06 elapsed=19:47:55 avg_step=1.33s recent_step=1.36s eta=17:12:29 +2026-08-25 23:43:40 | INFO | train rank=0 | step=53510/100000 progress=53.5% loss=0.046112 lr=4.833031e-06 elapsed=19:48:10 avg_step=1.33s recent_step=1.50s eta=17:12:17 +2026-08-25 23:43:56 | INFO | train rank=0 | step=53520/100000 progress=53.5% loss=0.045677 lr=4.831378e-06 elapsed=19:48:26 avg_step=1.33s recent_step=1.61s eta=17:12:06 +2026-08-25 23:44:08 | INFO | train rank=0 | step=53530/100000 progress=53.5% loss=0.049393 lr=4.829726e-06 elapsed=19:48:38 avg_step=1.33s recent_step=1.22s eta=17:11:52 +2026-08-25 23:44:23 | INFO | train rank=0 | step=53540/100000 progress=53.5% loss=0.052773 lr=4.828073e-06 elapsed=19:48:53 avg_step=1.33s recent_step=1.47s eta=17:11:40 +2026-08-25 23:44:38 | INFO | train rank=0 | step=53550/100000 progress=53.5% loss=0.036606 lr=4.826421e-06 elapsed=19:49:08 avg_step=1.33s recent_step=1.51s eta=17:11:28 +2026-08-25 23:44:51 | INFO | train rank=0 | step=53560/100000 progress=53.6% loss=0.047287 lr=4.824768e-06 elapsed=19:49:21 avg_step=1.33s recent_step=1.29s eta=17:11:14 +2026-08-25 23:45:02 | INFO | train rank=0 | step=53570/100000 progress=53.6% loss=0.075104 lr=4.823116e-06 elapsed=19:49:32 avg_step=1.33s recent_step=1.14s eta=17:10:59 +2026-08-25 23:45:21 | INFO | train rank=0 | step=53580/100000 progress=53.6% loss=0.042770 lr=4.821463e-06 elapsed=19:49:51 avg_step=1.33s recent_step=1.90s eta=17:10:51 +2026-08-25 23:45:33 | INFO | train rank=0 | step=53590/100000 progress=53.6% loss=0.048256 lr=4.819811e-06 elapsed=19:50:03 avg_step=1.33s recent_step=1.18s eta=17:10:36 +2026-08-25 23:45:47 | INFO | train rank=0 | step=53600/100000 progress=53.6% loss=0.081818 lr=4.818158e-06 elapsed=19:50:16 avg_step=1.33s recent_step=1.35s eta=17:10:23 +2026-08-25 23:46:00 | INFO | train rank=0 | step=53610/100000 progress=53.6% loss=0.078827 lr=4.816506e-06 elapsed=19:50:29 avg_step=1.33s recent_step=1.30s eta=17:10:09 +2026-08-25 23:46:11 | INFO | train rank=0 | step=53620/100000 progress=53.6% loss=0.058319 lr=4.814854e-06 elapsed=19:50:41 avg_step=1.33s recent_step=1.15s eta=17:09:55 +2026-08-25 23:46:30 | INFO | train rank=0 | step=53630/100000 progress=53.6% loss=0.071625 lr=4.813201e-06 elapsed=19:51:00 avg_step=1.33s recent_step=1.91s eta=17:09:46 +2026-08-25 23:46:41 | INFO | train rank=0 | step=53640/100000 progress=53.6% loss=0.061157 lr=4.811549e-06 elapsed=19:51:11 avg_step=1.33s recent_step=1.13s eta=17:09:31 +2026-08-25 23:46:53 | INFO | train rank=0 | step=53650/100000 progress=53.6% loss=0.052856 lr=4.809897e-06 elapsed=19:51:23 avg_step=1.33s recent_step=1.16s eta=17:09:16 +2026-08-25 23:47:07 | INFO | train rank=0 | step=53660/100000 progress=53.7% loss=0.053604 lr=4.808245e-06 elapsed=19:51:37 avg_step=1.33s recent_step=1.43s eta=17:09:04 +2026-08-25 23:47:24 | INFO | train rank=0 | step=53670/100000 progress=53.7% loss=0.057251 lr=4.806592e-06 elapsed=19:51:54 avg_step=1.33s recent_step=1.65s eta=17:08:53 +2026-08-25 23:47:35 | INFO | train rank=0 | step=53680/100000 progress=53.7% loss=0.038742 lr=4.804940e-06 elapsed=19:52:05 avg_step=1.33s recent_step=1.15s eta=17:08:38 +2026-08-25 23:47:49 | INFO | train rank=0 | step=53690/100000 progress=53.7% loss=0.047989 lr=4.803288e-06 elapsed=19:52:19 avg_step=1.33s recent_step=1.34s eta=17:08:25 +2026-08-25 23:48:00 | INFO | train rank=0 | step=53700/100000 progress=53.7% loss=0.048996 lr=4.801636e-06 elapsed=19:52:30 avg_step=1.33s recent_step=1.15s eta=17:08:10 +2026-08-25 23:48:15 | INFO | train rank=0 | step=53710/100000 progress=53.7% loss=0.041824 lr=4.799984e-06 elapsed=19:52:45 avg_step=1.33s recent_step=1.52s eta=17:07:59 +2026-08-25 23:48:27 | INFO | train rank=0 | step=53720/100000 progress=53.7% loss=0.068970 lr=4.798331e-06 elapsed=19:52:57 avg_step=1.33s recent_step=1.19s eta=17:07:44 +2026-08-25 23:48:39 | INFO | train rank=0 | step=53730/100000 progress=53.7% loss=0.080292 lr=4.796679e-06 elapsed=19:53:09 avg_step=1.33s recent_step=1.14s eta=17:07:29 +2026-08-25 23:48:51 | INFO | train rank=0 | step=53740/100000 progress=53.7% loss=0.055939 lr=4.795027e-06 elapsed=19:53:21 avg_step=1.33s recent_step=1.24s eta=17:07:15 +2026-08-25 23:49:05 | INFO | train rank=0 | step=53750/100000 progress=53.8% loss=0.025787 lr=4.793375e-06 elapsed=19:53:35 avg_step=1.33s recent_step=1.37s eta=17:07:02 +2026-08-25 23:49:21 | INFO | train rank=0 | step=53760/100000 progress=53.8% loss=0.048874 lr=4.791723e-06 elapsed=19:53:50 avg_step=1.33s recent_step=1.56s eta=17:06:51 +2026-08-25 23:49:35 | INFO | train rank=0 | step=53770/100000 progress=53.8% loss=0.060577 lr=4.790071e-06 elapsed=19:54:05 avg_step=1.33s recent_step=1.44s eta=17:06:38 +2026-08-25 23:49:47 | INFO | train rank=0 | step=53780/100000 progress=53.8% loss=0.040924 lr=4.788419e-06 elapsed=19:54:17 avg_step=1.33s recent_step=1.17s eta=17:06:24 +2026-08-25 23:49:59 | INFO | train rank=0 | step=53790/100000 progress=53.8% loss=0.044052 lr=4.786767e-06 elapsed=19:54:29 avg_step=1.33s recent_step=1.23s eta=17:06:09 +2026-08-25 23:50:16 | INFO | train rank=0 | step=53800/100000 progress=53.8% loss=0.041588 lr=4.785115e-06 elapsed=19:54:46 avg_step=1.33s recent_step=1.70s eta=17:05:59 +2026-08-25 23:50:30 | INFO | train rank=0 | step=53810/100000 progress=53.8% loss=0.077515 lr=4.783463e-06 elapsed=19:55:00 avg_step=1.33s recent_step=1.40s eta=17:05:46 +2026-08-25 23:50:41 | INFO | train rank=0 | step=53820/100000 progress=53.8% loss=0.066956 lr=4.781811e-06 elapsed=19:55:11 avg_step=1.33s recent_step=1.14s eta=17:05:31 +2026-08-25 23:50:56 | INFO | train rank=0 | step=53830/100000 progress=53.8% loss=0.032532 lr=4.780159e-06 elapsed=19:55:26 avg_step=1.33s recent_step=1.43s eta=17:05:19 +2026-08-25 23:51:08 | INFO | train rank=0 | step=53840/100000 progress=53.8% loss=0.047302 lr=4.778508e-06 elapsed=19:55:38 avg_step=1.33s recent_step=1.22s eta=17:05:05 +2026-08-25 23:51:21 | INFO | train rank=0 | step=53850/100000 progress=53.8% loss=0.048851 lr=4.776856e-06 elapsed=19:55:51 avg_step=1.33s recent_step=1.29s eta=17:04:51 +2026-08-25 23:51:35 | INFO | train rank=0 | step=53860/100000 progress=53.9% loss=0.029831 lr=4.775204e-06 elapsed=19:56:05 avg_step=1.33s recent_step=1.41s eta=17:04:38 +2026-08-25 23:51:48 | INFO | train rank=0 | step=53870/100000 progress=53.9% loss=0.044022 lr=4.773552e-06 elapsed=19:56:18 avg_step=1.33s recent_step=1.32s eta=17:04:25 +2026-08-25 23:52:00 | INFO | train rank=0 | step=53880/100000 progress=53.9% loss=0.051567 lr=4.771900e-06 elapsed=19:56:30 avg_step=1.33s recent_step=1.16s eta=17:04:10 +2026-08-25 23:52:12 | INFO | train rank=0 | step=53890/100000 progress=53.9% loss=0.044662 lr=4.770249e-06 elapsed=19:56:42 avg_step=1.33s recent_step=1.23s eta=17:03:56 +2026-08-25 23:52:24 | INFO | train rank=0 | step=53900/100000 progress=53.9% loss=0.057358 lr=4.768597e-06 elapsed=19:56:54 avg_step=1.33s recent_step=1.16s eta=17:03:41 +2026-08-25 23:52:35 | INFO | train rank=0 | step=53910/100000 progress=53.9% loss=0.047485 lr=4.766945e-06 elapsed=19:57:05 avg_step=1.33s recent_step=1.17s eta=17:03:26 +2026-08-25 23:52:48 | INFO | train rank=0 | step=53920/100000 progress=53.9% loss=0.058426 lr=4.765294e-06 elapsed=19:57:18 avg_step=1.33s recent_step=1.25s eta=17:03:12 +2026-08-25 23:53:03 | INFO | train rank=0 | step=53930/100000 progress=53.9% loss=0.081665 lr=4.763642e-06 elapsed=19:57:33 avg_step=1.33s recent_step=1.50s eta=17:03:01 +2026-08-25 23:53:15 | INFO | train rank=0 | step=53940/100000 progress=53.9% loss=0.050987 lr=4.761990e-06 elapsed=19:57:44 avg_step=1.33s recent_step=1.17s eta=17:02:46 +2026-08-25 23:53:28 | INFO | train rank=0 | step=53950/100000 progress=53.9% loss=0.047226 lr=4.760339e-06 elapsed=19:57:58 avg_step=1.33s recent_step=1.31s eta=17:02:32 +2026-08-25 23:53:42 | INFO | train rank=0 | step=53960/100000 progress=54.0% loss=0.075897 lr=4.758687e-06 elapsed=19:58:12 avg_step=1.33s recent_step=1.45s eta=17:02:20 +2026-08-25 23:53:55 | INFO | train rank=0 | step=53970/100000 progress=54.0% loss=0.125916 lr=4.757036e-06 elapsed=19:58:25 avg_step=1.33s recent_step=1.27s eta=17:02:06 +2026-08-25 23:54:10 | INFO | train rank=0 | step=53980/100000 progress=54.0% loss=0.115234 lr=4.755384e-06 elapsed=19:58:40 avg_step=1.33s recent_step=1.55s eta=17:01:55 +2026-08-25 23:54:24 | INFO | train rank=0 | step=53990/100000 progress=54.0% loss=0.050842 lr=4.753733e-06 elapsed=19:58:54 avg_step=1.33s recent_step=1.38s eta=17:01:42 +2026-08-25 23:54:37 | INFO | train rank=0 | step=54000/100000 progress=54.0% loss=0.039108 lr=4.752081e-06 elapsed=19:59:06 avg_step=1.33s recent_step=1.25s eta=17:01:28 +2026-08-25 23:54:52 | INFO | train rank=0 | step=54010/100000 progress=54.0% loss=0.047958 lr=4.750430e-06 elapsed=19:59:22 avg_step=1.33s recent_step=1.54s eta=17:01:16 +2026-08-25 23:55:05 | INFO | train rank=0 | step=54020/100000 progress=54.0% loss=0.053001 lr=4.748778e-06 elapsed=19:59:35 avg_step=1.33s recent_step=1.29s eta=17:01:02 +2026-08-25 23:55:20 | INFO | train rank=0 | step=54030/100000 progress=54.0% loss=0.042099 lr=4.747127e-06 elapsed=19:59:50 avg_step=1.33s recent_step=1.51s eta=17:00:51 +2026-08-25 23:55:32 | INFO | train rank=0 | step=54040/100000 progress=54.0% loss=0.030052 lr=4.745476e-06 elapsed=20:00:02 avg_step=1.33s recent_step=1.23s eta=17:00:36 +2026-08-25 23:55:46 | INFO | train rank=0 | step=54050/100000 progress=54.0% loss=0.049316 lr=4.743824e-06 elapsed=20:00:16 avg_step=1.33s recent_step=1.41s eta=17:00:24 +2026-08-25 23:55:59 | INFO | train rank=0 | step=54060/100000 progress=54.1% loss=0.061707 lr=4.742173e-06 elapsed=20:00:29 avg_step=1.33s recent_step=1.32s eta=17:00:10 +2026-08-25 23:56:14 | INFO | train rank=0 | step=54070/100000 progress=54.1% loss=0.042603 lr=4.740522e-06 elapsed=20:00:44 avg_step=1.33s recent_step=1.48s eta=16:59:58 +2026-08-25 23:56:29 | INFO | train rank=0 | step=54080/100000 progress=54.1% loss=0.096252 lr=4.738871e-06 elapsed=20:00:59 avg_step=1.33s recent_step=1.51s eta=16:59:46 +2026-08-25 23:56:41 | INFO | train rank=0 | step=54090/100000 progress=54.1% loss=0.049042 lr=4.737219e-06 elapsed=20:01:11 avg_step=1.33s recent_step=1.14s eta=16:59:31 +2026-08-25 23:56:54 | INFO | train rank=0 | step=54100/100000 progress=54.1% loss=0.053604 lr=4.735568e-06 elapsed=20:01:23 avg_step=1.33s recent_step=1.28s eta=16:59:18 +2026-08-25 23:57:08 | INFO | train rank=0 | step=54110/100000 progress=54.1% loss=0.055748 lr=4.733917e-06 elapsed=20:01:38 avg_step=1.33s recent_step=1.44s eta=16:59:05 +2026-08-25 23:57:22 | INFO | train rank=0 | step=54120/100000 progress=54.1% loss=0.029602 lr=4.732266e-06 elapsed=20:01:51 avg_step=1.33s recent_step=1.36s eta=16:58:52 +2026-08-25 23:57:36 | INFO | train rank=0 | step=54130/100000 progress=54.1% loss=0.036789 lr=4.730615e-06 elapsed=20:02:06 avg_step=1.33s recent_step=1.41s eta=16:58:39 +2026-08-25 23:57:47 | INFO | train rank=0 | step=54140/100000 progress=54.1% loss=0.070496 lr=4.728964e-06 elapsed=20:02:17 avg_step=1.33s recent_step=1.15s eta=16:58:25 +2026-08-25 23:58:00 | INFO | train rank=0 | step=54150/100000 progress=54.1% loss=0.081024 lr=4.727313e-06 elapsed=20:02:30 avg_step=1.33s recent_step=1.31s eta=16:58:11 +2026-08-25 23:58:18 | INFO | train rank=0 | step=54160/100000 progress=54.2% loss=0.044708 lr=4.725662e-06 elapsed=20:02:48 avg_step=1.33s recent_step=1.79s eta=16:58:02 +2026-08-25 23:58:32 | INFO | train rank=0 | step=54170/100000 progress=54.2% loss=0.046600 lr=4.724011e-06 elapsed=20:03:02 avg_step=1.33s recent_step=1.42s eta=16:57:49 +2026-08-25 23:58:45 | INFO | train rank=0 | step=54180/100000 progress=54.2% loss=0.065704 lr=4.722360e-06 elapsed=20:03:15 avg_step=1.33s recent_step=1.25s eta=16:57:35 +2026-08-25 23:58:56 | INFO | train rank=0 | step=54190/100000 progress=54.2% loss=0.054535 lr=4.720709e-06 elapsed=20:03:26 avg_step=1.33s recent_step=1.16s eta=16:57:20 +2026-08-25 23:59:11 | INFO | train rank=0 | step=54200/100000 progress=54.2% loss=0.036392 lr=4.719058e-06 elapsed=20:03:41 avg_step=1.33s recent_step=1.44s eta=16:57:08 +2026-08-25 23:59:22 | INFO | train rank=0 | step=54210/100000 progress=54.2% loss=0.064636 lr=4.717407e-06 elapsed=20:03:52 avg_step=1.33s recent_step=1.12s eta=16:56:53 +2026-08-25 23:59:37 | INFO | train rank=0 | step=54220/100000 progress=54.2% loss=0.046967 lr=4.715757e-06 elapsed=20:04:07 avg_step=1.33s recent_step=1.51s eta=16:56:41 +2026-08-25 23:59:49 | INFO | train rank=0 | step=54230/100000 progress=54.2% loss=0.057022 lr=4.714106e-06 elapsed=20:04:19 avg_step=1.33s recent_step=1.19s eta=16:56:26 +2026-08-26 00:00:01 | INFO | train rank=0 | step=54240/100000 progress=54.2% loss=0.079102 lr=4.712455e-06 elapsed=20:04:31 avg_step=1.33s recent_step=1.18s eta=16:56:12 +2026-08-26 00:00:15 | INFO | train rank=0 | step=54250/100000 progress=54.2% loss=0.040787 lr=4.710804e-06 elapsed=20:04:45 avg_step=1.33s recent_step=1.45s eta=16:55:59 +2026-08-26 00:00:28 | INFO | train rank=0 | step=54260/100000 progress=54.3% loss=0.088135 lr=4.709154e-06 elapsed=20:04:57 avg_step=1.33s recent_step=1.23s eta=16:55:45 +2026-08-26 00:00:40 | INFO | train rank=0 | step=54270/100000 progress=54.3% loss=0.048141 lr=4.707503e-06 elapsed=20:05:10 avg_step=1.33s recent_step=1.23s eta=16:55:31 +2026-08-26 00:00:52 | INFO | train rank=0 | step=54280/100000 progress=54.3% loss=0.033478 lr=4.705852e-06 elapsed=20:05:22 avg_step=1.33s recent_step=1.20s eta=16:55:16 +2026-08-26 00:01:06 | INFO | train rank=0 | step=54290/100000 progress=54.3% loss=0.040314 lr=4.704202e-06 elapsed=20:05:36 avg_step=1.33s recent_step=1.46s eta=16:55:04 +2026-08-26 00:01:18 | INFO | train rank=0 | step=54300/100000 progress=54.3% loss=0.029572 lr=4.702551e-06 elapsed=20:05:48 avg_step=1.33s recent_step=1.18s eta=16:54:49 +2026-08-26 00:01:33 | INFO | train rank=0 | step=54310/100000 progress=54.3% loss=0.056549 lr=4.700901e-06 elapsed=20:06:03 avg_step=1.33s recent_step=1.54s eta=16:54:38 +2026-08-26 00:01:49 | INFO | train rank=0 | step=54320/100000 progress=54.3% loss=0.043289 lr=4.699250e-06 elapsed=20:06:19 avg_step=1.33s recent_step=1.57s eta=16:54:27 +2026-08-26 00:02:03 | INFO | train rank=0 | step=54330/100000 progress=54.3% loss=0.033424 lr=4.697600e-06 elapsed=20:06:33 avg_step=1.33s recent_step=1.40s eta=16:54:14 +2026-08-26 00:02:17 | INFO | train rank=0 | step=54340/100000 progress=54.3% loss=0.042053 lr=4.695949e-06 elapsed=20:06:47 avg_step=1.33s recent_step=1.36s eta=16:54:01 +2026-08-26 00:02:32 | INFO | train rank=0 | step=54350/100000 progress=54.4% loss=0.056152 lr=4.694299e-06 elapsed=20:07:02 avg_step=1.33s recent_step=1.54s eta=16:53:49 +2026-08-26 00:02:48 | INFO | train rank=0 | step=54360/100000 progress=54.4% loss=0.052734 lr=4.692649e-06 elapsed=20:07:18 avg_step=1.33s recent_step=1.56s eta=16:53:38 +2026-08-26 00:02:59 | INFO | train rank=0 | step=54370/100000 progress=54.4% loss=0.039398 lr=4.690998e-06 elapsed=20:07:29 avg_step=1.33s recent_step=1.13s eta=16:53:23 +2026-08-26 00:03:11 | INFO | train rank=0 | step=54380/100000 progress=54.4% loss=0.037338 lr=4.689348e-06 elapsed=20:07:41 avg_step=1.33s recent_step=1.20s eta=16:53:08 +2026-08-26 00:03:22 | INFO | train rank=0 | step=54390/100000 progress=54.4% loss=0.061371 lr=4.687698e-06 elapsed=20:07:52 avg_step=1.33s recent_step=1.13s eta=16:52:53 +2026-08-26 00:03:39 | INFO | train rank=0 | step=54400/100000 progress=54.4% loss=0.037903 lr=4.686047e-06 elapsed=20:08:08 avg_step=1.33s recent_step=1.61s eta=16:52:42 +2026-08-26 00:03:50 | INFO | train rank=0 | step=54410/100000 progress=54.4% loss=0.038708 lr=4.684397e-06 elapsed=20:08:20 avg_step=1.33s recent_step=1.14s eta=16:52:27 +2026-08-26 00:04:07 | INFO | train rank=0 | step=54420/100000 progress=54.4% loss=0.067001 lr=4.682747e-06 elapsed=20:08:37 avg_step=1.33s recent_step=1.72s eta=16:52:17 +2026-08-26 00:04:19 | INFO | train rank=0 | step=54430/100000 progress=54.4% loss=0.059631 lr=4.681097e-06 elapsed=20:08:49 avg_step=1.33s recent_step=1.19s eta=16:52:03 +2026-08-26 00:04:38 | INFO | train rank=0 | step=54440/100000 progress=54.4% loss=0.060791 lr=4.679447e-06 elapsed=20:09:08 avg_step=1.33s recent_step=1.90s eta=16:51:54 +2026-08-26 00:04:50 | INFO | train rank=0 | step=54450/100000 progress=54.4% loss=0.056519 lr=4.677797e-06 elapsed=20:09:20 avg_step=1.33s recent_step=1.16s eta=16:51:39 +2026-08-26 00:05:01 | INFO | train rank=0 | step=54460/100000 progress=54.5% loss=0.038040 lr=4.676147e-06 elapsed=20:09:31 avg_step=1.33s recent_step=1.14s eta=16:51:24 +2026-08-26 00:05:15 | INFO | train rank=0 | step=54470/100000 progress=54.5% loss=0.061554 lr=4.674497e-06 elapsed=20:09:45 avg_step=1.33s recent_step=1.39s eta=16:51:12 +2026-08-26 00:05:28 | INFO | train rank=0 | step=54480/100000 progress=54.5% loss=0.044128 lr=4.672847e-06 elapsed=20:09:58 avg_step=1.33s recent_step=1.33s eta=16:50:58 +2026-08-26 00:05:40 | INFO | train rank=0 | step=54490/100000 progress=54.5% loss=0.101501 lr=4.671197e-06 elapsed=20:10:10 avg_step=1.33s recent_step=1.15s eta=16:50:43 +2026-08-26 00:05:54 | INFO | train rank=0 | step=54500/100000 progress=54.5% loss=0.028206 lr=4.669547e-06 elapsed=20:10:24 avg_step=1.33s recent_step=1.42s eta=16:50:31 +2026-08-26 00:06:09 | INFO | train rank=0 | step=54510/100000 progress=54.5% loss=0.050674 lr=4.667897e-06 elapsed=20:10:39 avg_step=1.33s recent_step=1.51s eta=16:50:19 +2026-08-26 00:06:24 | INFO | train rank=0 | step=54520/100000 progress=54.5% loss=0.101227 lr=4.666247e-06 elapsed=20:10:53 avg_step=1.33s recent_step=1.44s eta=16:50:07 +2026-08-26 00:06:40 | INFO | train rank=0 | step=54530/100000 progress=54.5% loss=0.042786 lr=4.664598e-06 elapsed=20:11:10 avg_step=1.33s recent_step=1.63s eta=16:49:56 +2026-08-26 00:06:51 | INFO | train rank=0 | step=54540/100000 progress=54.5% loss=0.059784 lr=4.662948e-06 elapsed=20:11:21 avg_step=1.33s recent_step=1.14s eta=16:49:41 +2026-08-26 00:07:03 | INFO | train rank=0 | step=54550/100000 progress=54.5% loss=0.035095 lr=4.661298e-06 elapsed=20:11:33 avg_step=1.33s recent_step=1.15s eta=16:49:26 +2026-08-26 00:07:18 | INFO | train rank=0 | step=54560/100000 progress=54.6% loss=0.069992 lr=4.659649e-06 elapsed=20:11:48 avg_step=1.33s recent_step=1.52s eta=16:49:14 +2026-08-26 00:07:29 | INFO | train rank=0 | step=54570/100000 progress=54.6% loss=0.080261 lr=4.657999e-06 elapsed=20:11:59 avg_step=1.33s recent_step=1.11s eta=16:48:59 +2026-08-26 00:07:41 | INFO | train rank=0 | step=54580/100000 progress=54.6% loss=0.063049 lr=4.656349e-06 elapsed=20:12:11 avg_step=1.33s recent_step=1.18s eta=16:48:45 +2026-08-26 00:07:52 | INFO | train rank=0 | step=54590/100000 progress=54.6% loss=0.064941 lr=4.654700e-06 elapsed=20:12:22 avg_step=1.33s recent_step=1.16s eta=16:48:30 +2026-08-26 00:08:08 | INFO | train rank=0 | step=54600/100000 progress=54.6% loss=0.041412 lr=4.653050e-06 elapsed=20:12:38 avg_step=1.33s recent_step=1.52s eta=16:48:18 +2026-08-26 00:08:24 | INFO | train rank=0 | step=54610/100000 progress=54.6% loss=0.051056 lr=4.651401e-06 elapsed=20:12:54 avg_step=1.33s recent_step=1.63s eta=16:48:07 +2026-08-26 00:08:36 | INFO | train rank=0 | step=54620/100000 progress=54.6% loss=0.062134 lr=4.649751e-06 elapsed=20:13:05 avg_step=1.33s recent_step=1.16s eta=16:47:52 +2026-08-26 00:08:48 | INFO | train rank=0 | step=54630/100000 progress=54.6% loss=0.040474 lr=4.648102e-06 elapsed=20:13:17 avg_step=1.33s recent_step=1.20s eta=16:47:38 +2026-08-26 00:09:04 | INFO | train rank=0 | step=54640/100000 progress=54.6% loss=0.066040 lr=4.646453e-06 elapsed=20:13:34 avg_step=1.33s recent_step=1.61s eta=16:47:27 +2026-08-26 00:09:15 | INFO | train rank=0 | step=54650/100000 progress=54.6% loss=0.085648 lr=4.644803e-06 elapsed=20:13:45 avg_step=1.33s recent_step=1.15s eta=16:47:12 +2026-08-26 00:09:29 | INFO | train rank=0 | step=54660/100000 progress=54.7% loss=0.037369 lr=4.643154e-06 elapsed=20:13:59 avg_step=1.33s recent_step=1.37s eta=16:46:59 +2026-08-26 00:09:43 | INFO | train rank=0 | step=54670/100000 progress=54.7% loss=0.034622 lr=4.641505e-06 elapsed=20:14:13 avg_step=1.33s recent_step=1.40s eta=16:46:46 +2026-08-26 00:09:57 | INFO | train rank=0 | step=54680/100000 progress=54.7% loss=0.033981 lr=4.639856e-06 elapsed=20:14:27 avg_step=1.33s recent_step=1.43s eta=16:46:34 +2026-08-26 00:10:09 | INFO | train rank=0 | step=54690/100000 progress=54.7% loss=0.099426 lr=4.638207e-06 elapsed=20:14:39 avg_step=1.33s recent_step=1.17s eta=16:46:19 +2026-08-26 00:10:24 | INFO | train rank=0 | step=54700/100000 progress=54.7% loss=0.076569 lr=4.636557e-06 elapsed=20:14:54 avg_step=1.33s recent_step=1.49s eta=16:46:07 +2026-08-26 00:10:39 | INFO | train rank=0 | step=54710/100000 progress=54.7% loss=0.042778 lr=4.634908e-06 elapsed=20:15:09 avg_step=1.33s recent_step=1.50s eta=16:45:55 +2026-08-26 00:10:50 | INFO | train rank=0 | step=54720/100000 progress=54.7% loss=0.084106 lr=4.633259e-06 elapsed=20:15:20 avg_step=1.33s recent_step=1.13s eta=16:45:40 +2026-08-26 00:11:06 | INFO | train rank=0 | step=54730/100000 progress=54.7% loss=0.031059 lr=4.631610e-06 elapsed=20:15:36 avg_step=1.33s recent_step=1.62s eta=16:45:29 +2026-08-26 00:11:18 | INFO | train rank=0 | step=54740/100000 progress=54.7% loss=0.057343 lr=4.629961e-06 elapsed=20:15:48 avg_step=1.33s recent_step=1.17s eta=16:45:15 +2026-08-26 00:11:34 | INFO | train rank=0 | step=54750/100000 progress=54.8% loss=0.052780 lr=4.628312e-06 elapsed=20:16:04 avg_step=1.33s recent_step=1.56s eta=16:45:03 +2026-08-26 00:11:48 | INFO | train rank=0 | step=54760/100000 progress=54.8% loss=0.054260 lr=4.626664e-06 elapsed=20:16:18 avg_step=1.33s recent_step=1.47s eta=16:44:51 +2026-08-26 00:12:00 | INFO | train rank=0 | step=54770/100000 progress=54.8% loss=0.045746 lr=4.625015e-06 elapsed=20:16:30 avg_step=1.33s recent_step=1.19s eta=16:44:36 +2026-08-26 00:12:12 | INFO | train rank=0 | step=54780/100000 progress=54.8% loss=0.036530 lr=4.623366e-06 elapsed=20:16:42 avg_step=1.33s recent_step=1.19s eta=16:44:22 +2026-08-26 00:12:24 | INFO | train rank=0 | step=54790/100000 progress=54.8% loss=0.049408 lr=4.621717e-06 elapsed=20:16:54 avg_step=1.33s recent_step=1.23s eta=16:44:08 +2026-08-26 00:12:36 | INFO | train rank=0 | step=54800/100000 progress=54.8% loss=0.065521 lr=4.620069e-06 elapsed=20:17:05 avg_step=1.33s recent_step=1.12s eta=16:43:53 +2026-08-26 00:12:50 | INFO | train rank=0 | step=54810/100000 progress=54.8% loss=0.043579 lr=4.618420e-06 elapsed=20:17:20 avg_step=1.33s recent_step=1.44s eta=16:43:40 +2026-08-26 00:13:02 | INFO | train rank=0 | step=54820/100000 progress=54.8% loss=0.084045 lr=4.616771e-06 elapsed=20:17:32 avg_step=1.33s recent_step=1.17s eta=16:43:25 +2026-08-26 00:13:13 | INFO | train rank=0 | step=54830/100000 progress=54.8% loss=0.059540 lr=4.615123e-06 elapsed=20:17:43 avg_step=1.33s recent_step=1.14s eta=16:43:11 +2026-08-26 00:13:30 | INFO | train rank=0 | step=54840/100000 progress=54.8% loss=0.034424 lr=4.613474e-06 elapsed=20:18:00 avg_step=1.33s recent_step=1.71s eta=16:43:00 +2026-08-26 00:13:42 | INFO | train rank=0 | step=54850/100000 progress=54.9% loss=0.059692 lr=4.611826e-06 elapsed=20:18:12 avg_step=1.33s recent_step=1.20s eta=16:42:46 +2026-08-26 00:13:57 | INFO | train rank=0 | step=54860/100000 progress=54.9% loss=0.031921 lr=4.610177e-06 elapsed=20:18:27 avg_step=1.33s recent_step=1.49s eta=16:42:34 +2026-08-26 00:14:11 | INFO | train rank=0 | step=54870/100000 progress=54.9% loss=0.041412 lr=4.608529e-06 elapsed=20:18:41 avg_step=1.33s recent_step=1.40s eta=16:42:21 +2026-08-26 00:14:28 | INFO | train rank=0 | step=54880/100000 progress=54.9% loss=0.070953 lr=4.606880e-06 elapsed=20:18:58 avg_step=1.33s recent_step=1.66s eta=16:42:11 +2026-08-26 00:14:40 | INFO | train rank=0 | step=54890/100000 progress=54.9% loss=0.028366 lr=4.605232e-06 elapsed=20:19:10 avg_step=1.33s recent_step=1.21s eta=16:41:56 +2026-08-26 00:14:52 | INFO | train rank=0 | step=54900/100000 progress=54.9% loss=0.034805 lr=4.603584e-06 elapsed=20:19:22 avg_step=1.33s recent_step=1.24s eta=16:41:42 +2026-08-26 00:15:06 | INFO | train rank=0 | step=54910/100000 progress=54.9% loss=0.047165 lr=4.601935e-06 elapsed=20:19:36 avg_step=1.33s recent_step=1.42s eta=16:41:30 +2026-08-26 00:15:26 | INFO | train rank=0 | step=54920/100000 progress=54.9% loss=0.027870 lr=4.600287e-06 elapsed=20:19:56 avg_step=1.33s recent_step=1.96s eta=16:41:21 +2026-08-26 00:15:37 | INFO | train rank=0 | step=54930/100000 progress=54.9% loss=0.040436 lr=4.598639e-06 elapsed=20:20:07 avg_step=1.33s recent_step=1.14s eta=16:41:06 +2026-08-26 00:15:52 | INFO | train rank=0 | step=54940/100000 progress=54.9% loss=0.037506 lr=4.596991e-06 elapsed=20:20:22 avg_step=1.33s recent_step=1.49s eta=16:40:54 +2026-08-26 00:16:04 | INFO | train rank=0 | step=54950/100000 progress=54.9% loss=0.055206 lr=4.595343e-06 elapsed=20:20:34 avg_step=1.33s recent_step=1.22s eta=16:40:40 +2026-08-26 00:16:20 | INFO | train rank=0 | step=54960/100000 progress=55.0% loss=0.049469 lr=4.593695e-06 elapsed=20:20:50 avg_step=1.33s recent_step=1.51s eta=16:40:28 +2026-08-26 00:16:32 | INFO | train rank=0 | step=54970/100000 progress=55.0% loss=0.069702 lr=4.592047e-06 elapsed=20:21:01 avg_step=1.33s recent_step=1.19s eta=16:40:14 +2026-08-26 00:16:44 | INFO | train rank=0 | step=54980/100000 progress=55.0% loss=0.045975 lr=4.590399e-06 elapsed=20:21:14 avg_step=1.33s recent_step=1.23s eta=16:40:00 +2026-08-26 00:16:55 | INFO | train rank=0 | step=54990/100000 progress=55.0% loss=0.037590 lr=4.588751e-06 elapsed=20:21:25 avg_step=1.33s recent_step=1.11s eta=16:39:44 +2026-08-26 00:17:11 | INFO | train rank=0 | step=55000/100000 progress=55.0% loss=0.045982 lr=4.587103e-06 elapsed=20:21:41 avg_step=1.33s recent_step=1.61s eta=16:39:33 +2026-08-26 00:17:27 | INFO | train rank=0 | step=55010/100000 progress=55.0% loss=0.051590 lr=4.585455e-06 elapsed=20:21:56 avg_step=1.33s recent_step=1.55s eta=16:39:22 +2026-08-26 00:17:38 | INFO | train rank=0 | step=55020/100000 progress=55.0% loss=0.036499 lr=4.583808e-06 elapsed=20:22:08 avg_step=1.33s recent_step=1.17s eta=16:39:07 +2026-08-26 00:17:50 | INFO | train rank=0 | step=55030/100000 progress=55.0% loss=0.048920 lr=4.582160e-06 elapsed=20:22:20 avg_step=1.33s recent_step=1.20s eta=16:38:53 +2026-08-26 00:18:07 | INFO | train rank=0 | step=55040/100000 progress=55.0% loss=0.043243 lr=4.580512e-06 elapsed=20:22:37 avg_step=1.33s recent_step=1.65s eta=16:38:42 +2026-08-26 00:18:18 | INFO | train rank=0 | step=55050/100000 progress=55.0% loss=0.040665 lr=4.578865e-06 elapsed=20:22:48 avg_step=1.33s recent_step=1.16s eta=16:38:27 +2026-08-26 00:18:34 | INFO | train rank=0 | step=55060/100000 progress=55.1% loss=0.095154 lr=4.577217e-06 elapsed=20:23:04 avg_step=1.33s recent_step=1.58s eta=16:38:16 +2026-08-26 00:18:47 | INFO | train rank=0 | step=55070/100000 progress=55.1% loss=0.046921 lr=4.575570e-06 elapsed=20:23:17 avg_step=1.33s recent_step=1.27s eta=16:38:02 +2026-08-26 00:19:06 | INFO | train rank=0 | step=55080/100000 progress=55.1% loss=0.037033 lr=4.573922e-06 elapsed=20:23:36 avg_step=1.33s recent_step=1.88s eta=16:37:53 +2026-08-26 00:19:17 | INFO | train rank=0 | step=55090/100000 progress=55.1% loss=0.088013 lr=4.572275e-06 elapsed=20:23:47 avg_step=1.33s recent_step=1.12s eta=16:37:38 +2026-08-26 00:19:32 | INFO | train rank=0 | step=55100/100000 progress=55.1% loss=0.105591 lr=4.570627e-06 elapsed=20:24:02 avg_step=1.33s recent_step=1.50s eta=16:37:26 +2026-08-26 00:19:43 | INFO | train rank=0 | step=55110/100000 progress=55.1% loss=0.039474 lr=4.568980e-06 elapsed=20:24:13 avg_step=1.33s recent_step=1.15s eta=16:37:11 +2026-08-26 00:20:04 | INFO | train rank=0 | step=55120/100000 progress=55.1% loss=0.055450 lr=4.567333e-06 elapsed=20:24:34 avg_step=1.33s recent_step=2.04s eta=16:37:04 +2026-08-26 00:20:15 | INFO | train rank=0 | step=55130/100000 progress=55.1% loss=0.055771 lr=4.565685e-06 elapsed=20:24:45 avg_step=1.33s recent_step=1.16s eta=16:36:49 +2026-08-26 00:20:27 | INFO | train rank=0 | step=55140/100000 progress=55.1% loss=0.048203 lr=4.564038e-06 elapsed=20:24:56 avg_step=1.33s recent_step=1.12s eta=16:36:34 +2026-08-26 00:20:40 | INFO | train rank=0 | step=55150/100000 progress=55.1% loss=0.074677 lr=4.562391e-06 elapsed=20:25:10 avg_step=1.33s recent_step=1.31s eta=16:36:21 +2026-08-26 00:20:52 | INFO | train rank=0 | step=55160/100000 progress=55.2% loss=0.037491 lr=4.560744e-06 elapsed=20:25:22 avg_step=1.33s recent_step=1.22s eta=16:36:06 +2026-08-26 00:21:06 | INFO | train rank=0 | step=55170/100000 progress=55.2% loss=0.042992 lr=4.559097e-06 elapsed=20:25:36 avg_step=1.33s recent_step=1.42s eta=16:35:54 +2026-08-26 00:21:20 | INFO | train rank=0 | step=55180/100000 progress=55.2% loss=0.034302 lr=4.557450e-06 elapsed=20:25:50 avg_step=1.33s recent_step=1.37s eta=16:35:41 +2026-08-26 00:21:36 | INFO | train rank=0 | step=55190/100000 progress=55.2% loss=0.042511 lr=4.555803e-06 elapsed=20:26:06 avg_step=1.33s recent_step=1.60s eta=16:35:30 +2026-08-26 00:21:49 | INFO | train rank=0 | step=55200/100000 progress=55.2% loss=0.043091 lr=4.554156e-06 elapsed=20:26:19 avg_step=1.33s recent_step=1.27s eta=16:35:16 +2026-08-26 00:22:01 | INFO | train rank=0 | step=55210/100000 progress=55.2% loss=0.065460 lr=4.552509e-06 elapsed=20:26:30 avg_step=1.33s recent_step=1.19s eta=16:35:01 +2026-08-26 00:22:15 | INFO | train rank=0 | step=55220/100000 progress=55.2% loss=0.035461 lr=4.550862e-06 elapsed=20:26:45 avg_step=1.33s recent_step=1.44s eta=16:34:49 +2026-08-26 00:22:26 | INFO | train rank=0 | step=55230/100000 progress=55.2% loss=0.091858 lr=4.549216e-06 elapsed=20:26:56 avg_step=1.33s recent_step=1.14s eta=16:34:34 +2026-08-26 00:22:43 | INFO | train rank=0 | step=55240/100000 progress=55.2% loss=0.050430 lr=4.547569e-06 elapsed=20:27:13 avg_step=1.33s recent_step=1.67s eta=16:34:23 +2026-08-26 00:22:55 | INFO | train rank=0 | step=55250/100000 progress=55.2% loss=0.064636 lr=4.545922e-06 elapsed=20:27:25 avg_step=1.33s recent_step=1.21s eta=16:34:09 +2026-08-26 00:23:06 | INFO | train rank=0 | step=55260/100000 progress=55.3% loss=0.041046 lr=4.544276e-06 elapsed=20:27:36 avg_step=1.33s recent_step=1.12s eta=16:33:54 +2026-08-26 00:23:18 | INFO | train rank=0 | step=55270/100000 progress=55.3% loss=0.077484 lr=4.542629e-06 elapsed=20:27:48 avg_step=1.33s recent_step=1.17s eta=16:33:39 +2026-08-26 00:23:31 | INFO | train rank=0 | step=55280/100000 progress=55.3% loss=0.061462 lr=4.540983e-06 elapsed=20:28:01 avg_step=1.33s recent_step=1.29s eta=16:33:26 +2026-08-26 00:23:43 | INFO | train rank=0 | step=55290/100000 progress=55.3% loss=0.059227 lr=4.539336e-06 elapsed=20:28:12 avg_step=1.33s recent_step=1.15s eta=16:33:11 +2026-08-26 00:23:56 | INFO | train rank=0 | step=55300/100000 progress=55.3% loss=0.042908 lr=4.537690e-06 elapsed=20:28:26 avg_step=1.33s recent_step=1.39s eta=16:32:58 +2026-08-26 00:24:08 | INFO | train rank=0 | step=55310/100000 progress=55.3% loss=0.061615 lr=4.536043e-06 elapsed=20:28:38 avg_step=1.33s recent_step=1.18s eta=16:32:44 +2026-08-26 00:24:24 | INFO | train rank=0 | step=55320/100000 progress=55.3% loss=0.023819 lr=4.534397e-06 elapsed=20:28:54 avg_step=1.33s recent_step=1.61s eta=16:32:32 +2026-08-26 00:24:37 | INFO | train rank=0 | step=55330/100000 progress=55.3% loss=0.041199 lr=4.532751e-06 elapsed=20:29:07 avg_step=1.33s recent_step=1.23s eta=16:32:18 +2026-08-26 00:24:48 | INFO | train rank=0 | step=55340/100000 progress=55.3% loss=0.052353 lr=4.531105e-06 elapsed=20:29:18 avg_step=1.33s recent_step=1.18s eta=16:32:04 +2026-08-26 00:25:08 | INFO | train rank=0 | step=55350/100000 progress=55.4% loss=0.041122 lr=4.529458e-06 elapsed=20:29:38 avg_step=1.33s recent_step=1.94s eta=16:31:55 +2026-08-26 00:25:24 | INFO | train rank=0 | step=55360/100000 progress=55.4% loss=0.045212 lr=4.527812e-06 elapsed=20:29:54 avg_step=1.33s recent_step=1.59s eta=16:31:44 +2026-08-26 00:25:35 | INFO | train rank=0 | step=55370/100000 progress=55.4% loss=0.042831 lr=4.526166e-06 elapsed=20:30:05 avg_step=1.33s recent_step=1.14s eta=16:31:29 +2026-08-26 00:25:48 | INFO | train rank=0 | step=55380/100000 progress=55.4% loss=0.047180 lr=4.524520e-06 elapsed=20:30:18 avg_step=1.33s recent_step=1.28s eta=16:31:15 +2026-08-26 00:26:00 | INFO | train rank=0 | step=55390/100000 progress=55.4% loss=0.055542 lr=4.522874e-06 elapsed=20:30:30 avg_step=1.33s recent_step=1.19s eta=16:31:01 +2026-08-26 00:26:15 | INFO | train rank=0 | step=55400/100000 progress=55.4% loss=0.043488 lr=4.521228e-06 elapsed=20:30:45 avg_step=1.33s recent_step=1.54s eta=16:30:49 +2026-08-26 00:26:27 | INFO | train rank=0 | step=55410/100000 progress=55.4% loss=0.031952 lr=4.519583e-06 elapsed=20:30:57 avg_step=1.33s recent_step=1.22s eta=16:30:35 +2026-08-26 00:26:39 | INFO | train rank=0 | step=55420/100000 progress=55.4% loss=0.066132 lr=4.517937e-06 elapsed=20:31:09 avg_step=1.33s recent_step=1.12s eta=16:30:20 +2026-08-26 00:26:52 | INFO | train rank=0 | step=55430/100000 progress=55.4% loss=0.055313 lr=4.516291e-06 elapsed=20:31:22 avg_step=1.33s recent_step=1.38s eta=16:30:07 +2026-08-26 00:27:08 | INFO | train rank=0 | step=55440/100000 progress=55.4% loss=0.116882 lr=4.514645e-06 elapsed=20:31:38 avg_step=1.33s recent_step=1.59s eta=16:29:56 +2026-08-26 00:27:25 | INFO | train rank=0 | step=55450/100000 progress=55.5% loss=0.041565 lr=4.513000e-06 elapsed=20:31:55 avg_step=1.33s recent_step=1.64s eta=16:29:45 +2026-08-26 00:27:37 | INFO | train rank=0 | step=55460/100000 progress=55.5% loss=0.048203 lr=4.511354e-06 elapsed=20:32:07 avg_step=1.33s recent_step=1.19s eta=16:29:30 +2026-08-26 00:27:50 | INFO | train rank=0 | step=55470/100000 progress=55.5% loss=0.035378 lr=4.509709e-06 elapsed=20:32:20 avg_step=1.33s recent_step=1.29s eta=16:29:17 +2026-08-26 00:28:04 | INFO | train rank=0 | step=55480/100000 progress=55.5% loss=0.049500 lr=4.508063e-06 elapsed=20:32:34 avg_step=1.33s recent_step=1.48s eta=16:29:05 +2026-08-26 00:28:16 | INFO | train rank=0 | step=55490/100000 progress=55.5% loss=0.050476 lr=4.506418e-06 elapsed=20:32:46 avg_step=1.33s recent_step=1.15s eta=16:28:50 +2026-08-26 00:28:29 | INFO | train rank=0 | step=55500/100000 progress=55.5% loss=0.063629 lr=4.504772e-06 elapsed=20:32:59 avg_step=1.33s recent_step=1.33s eta=16:28:37 +2026-08-26 00:28:44 | INFO | train rank=0 | step=55510/100000 progress=55.5% loss=0.063019 lr=4.503127e-06 elapsed=20:33:14 avg_step=1.33s recent_step=1.48s eta=16:28:24 +2026-08-26 00:29:00 | INFO | train rank=0 | step=55520/100000 progress=55.5% loss=0.078094 lr=4.501482e-06 elapsed=20:33:30 avg_step=1.33s recent_step=1.59s eta=16:28:13 +2026-08-26 00:29:11 | INFO | train rank=0 | step=55530/100000 progress=55.5% loss=0.054535 lr=4.499837e-06 elapsed=20:33:41 avg_step=1.33s recent_step=1.12s eta=16:27:58 +2026-08-26 00:29:25 | INFO | train rank=0 | step=55540/100000 progress=55.5% loss=0.045258 lr=4.498191e-06 elapsed=20:33:55 avg_step=1.33s recent_step=1.42s eta=16:27:45 +2026-08-26 00:29:39 | INFO | train rank=0 | step=55550/100000 progress=55.5% loss=0.052185 lr=4.496546e-06 elapsed=20:34:09 avg_step=1.33s recent_step=1.40s eta=16:27:33 +2026-08-26 00:29:51 | INFO | train rank=0 | step=55560/100000 progress=55.6% loss=0.040092 lr=4.494901e-06 elapsed=20:34:21 avg_step=1.33s recent_step=1.17s eta=16:27:18 +2026-08-26 00:30:09 | INFO | train rank=0 | step=55570/100000 progress=55.6% loss=0.074127 lr=4.493256e-06 elapsed=20:34:39 avg_step=1.33s recent_step=1.82s eta=16:27:09 +2026-08-26 00:30:21 | INFO | train rank=0 | step=55580/100000 progress=55.6% loss=0.038742 lr=4.491611e-06 elapsed=20:34:51 avg_step=1.33s recent_step=1.17s eta=16:26:54 +2026-08-26 00:30:32 | INFO | train rank=0 | step=55590/100000 progress=55.6% loss=0.041534 lr=4.489967e-06 elapsed=20:35:02 avg_step=1.33s recent_step=1.15s eta=16:26:39 +2026-08-26 00:30:49 | INFO | train rank=0 | step=55600/100000 progress=55.6% loss=0.048676 lr=4.488322e-06 elapsed=20:35:18 avg_step=1.33s recent_step=1.61s eta=16:26:28 +2026-08-26 00:31:01 | INFO | train rank=0 | step=55610/100000 progress=55.6% loss=0.043442 lr=4.486677e-06 elapsed=20:35:30 avg_step=1.33s recent_step=1.20s eta=16:26:14 +2026-08-26 00:31:16 | INFO | train rank=0 | step=55620/100000 progress=55.6% loss=0.103333 lr=4.485032e-06 elapsed=20:35:46 avg_step=1.33s recent_step=1.58s eta=16:26:02 +2026-08-26 00:31:28 | INFO | train rank=0 | step=55630/100000 progress=55.6% loss=0.051361 lr=4.483388e-06 elapsed=20:35:58 avg_step=1.33s recent_step=1.18s eta=16:25:48 +2026-08-26 00:31:40 | INFO | train rank=0 | step=55640/100000 progress=55.6% loss=0.040817 lr=4.481743e-06 elapsed=20:36:10 avg_step=1.33s recent_step=1.21s eta=16:25:33 +2026-08-26 00:31:56 | INFO | train rank=0 | step=55650/100000 progress=55.6% loss=0.048508 lr=4.480098e-06 elapsed=20:36:25 avg_step=1.33s recent_step=1.53s eta=16:25:22 +2026-08-26 00:32:10 | INFO | train rank=0 | step=55660/100000 progress=55.7% loss=0.051666 lr=4.478454e-06 elapsed=20:36:40 avg_step=1.33s recent_step=1.42s eta=16:25:09 +2026-08-26 00:32:22 | INFO | train rank=0 | step=55670/100000 progress=55.7% loss=0.036041 lr=4.476810e-06 elapsed=20:36:52 avg_step=1.33s recent_step=1.20s eta=16:24:55 +2026-08-26 00:32:33 | INFO | train rank=0 | step=55680/100000 progress=55.7% loss=0.034332 lr=4.475165e-06 elapsed=20:37:03 avg_step=1.33s recent_step=1.16s eta=16:24:40 +2026-08-26 00:32:46 | INFO | train rank=0 | step=55690/100000 progress=55.7% loss=0.053848 lr=4.473521e-06 elapsed=20:37:16 avg_step=1.33s recent_step=1.30s eta=16:24:26 +2026-08-26 00:33:00 | INFO | train rank=0 | step=55700/100000 progress=55.7% loss=0.062683 lr=4.471877e-06 elapsed=20:37:30 avg_step=1.33s recent_step=1.37s eta=16:24:13 +2026-08-26 00:33:14 | INFO | train rank=0 | step=55710/100000 progress=55.7% loss=0.054825 lr=4.470232e-06 elapsed=20:37:44 avg_step=1.33s recent_step=1.41s eta=16:24:01 +2026-08-26 00:33:26 | INFO | train rank=0 | step=55720/100000 progress=55.7% loss=0.040939 lr=4.468588e-06 elapsed=20:37:56 avg_step=1.33s recent_step=1.19s eta=16:23:46 +2026-08-26 00:33:40 | INFO | train rank=0 | step=55730/100000 progress=55.7% loss=0.040024 lr=4.466944e-06 elapsed=20:38:10 avg_step=1.33s recent_step=1.38s eta=16:23:33 +2026-08-26 00:33:53 | INFO | train rank=0 | step=55740/100000 progress=55.7% loss=0.039490 lr=4.465300e-06 elapsed=20:38:23 avg_step=1.33s recent_step=1.33s eta=16:23:20 +2026-08-26 00:34:09 | INFO | train rank=0 | step=55750/100000 progress=55.8% loss=0.040894 lr=4.463656e-06 elapsed=20:38:39 avg_step=1.33s recent_step=1.56s eta=16:23:08 +2026-08-26 00:34:26 | INFO | train rank=0 | step=55760/100000 progress=55.8% loss=0.051414 lr=4.462012e-06 elapsed=20:38:55 avg_step=1.33s recent_step=1.68s eta=16:22:58 +2026-08-26 00:34:40 | INFO | train rank=0 | step=55770/100000 progress=55.8% loss=0.094727 lr=4.460369e-06 elapsed=20:39:10 avg_step=1.33s recent_step=1.43s eta=16:22:45 +2026-08-26 00:34:55 | INFO | train rank=0 | step=55780/100000 progress=55.8% loss=0.085144 lr=4.458725e-06 elapsed=20:39:24 avg_step=1.33s recent_step=1.47s eta=16:22:33 +2026-08-26 00:35:06 | INFO | train rank=0 | step=55790/100000 progress=55.8% loss=0.047958 lr=4.457081e-06 elapsed=20:39:36 avg_step=1.33s recent_step=1.16s eta=16:22:18 +2026-08-26 00:35:18 | INFO | train rank=0 | step=55800/100000 progress=55.8% loss=0.029846 lr=4.455437e-06 elapsed=20:39:48 avg_step=1.33s recent_step=1.20s eta=16:22:04 +2026-08-26 00:35:30 | INFO | train rank=0 | step=55810/100000 progress=55.8% loss=0.066681 lr=4.453794e-06 elapsed=20:40:00 avg_step=1.33s recent_step=1.19s eta=16:21:49 +2026-08-26 00:35:45 | INFO | train rank=0 | step=55820/100000 progress=55.8% loss=0.038330 lr=4.452150e-06 elapsed=20:40:15 avg_step=1.33s recent_step=1.51s eta=16:21:37 +2026-08-26 00:35:57 | INFO | train rank=0 | step=55830/100000 progress=55.8% loss=0.035461 lr=4.450507e-06 elapsed=20:40:27 avg_step=1.33s recent_step=1.22s eta=16:21:23 +2026-08-26 00:36:12 | INFO | train rank=0 | step=55840/100000 progress=55.8% loss=0.105347 lr=4.448863e-06 elapsed=20:40:42 avg_step=1.33s recent_step=1.51s eta=16:21:11 +2026-08-26 00:36:24 | INFO | train rank=0 | step=55850/100000 progress=55.9% loss=0.058182 lr=4.447220e-06 elapsed=20:40:54 avg_step=1.33s recent_step=1.16s eta=16:20:57 +2026-08-26 00:36:38 | INFO | train rank=0 | step=55860/100000 progress=55.9% loss=0.111206 lr=4.445577e-06 elapsed=20:41:08 avg_step=1.33s recent_step=1.39s eta=16:20:44 +2026-08-26 00:36:52 | INFO | train rank=0 | step=55870/100000 progress=55.9% loss=0.044094 lr=4.443933e-06 elapsed=20:41:22 avg_step=1.33s recent_step=1.43s eta=16:20:31 +2026-08-26 00:37:04 | INFO | train rank=0 | step=55880/100000 progress=55.9% loss=0.045151 lr=4.442290e-06 elapsed=20:41:34 avg_step=1.33s recent_step=1.16s eta=16:20:16 +2026-08-26 00:37:19 | INFO | train rank=0 | step=55890/100000 progress=55.9% loss=0.023911 lr=4.440647e-06 elapsed=20:41:49 avg_step=1.33s recent_step=1.48s eta=16:20:04 +2026-08-26 00:37:31 | INFO | train rank=0 | step=55900/100000 progress=55.9% loss=0.051743 lr=4.439004e-06 elapsed=20:42:01 avg_step=1.33s recent_step=1.22s eta=16:19:50 +2026-08-26 00:37:45 | INFO | train rank=0 | step=55910/100000 progress=55.9% loss=0.059082 lr=4.437361e-06 elapsed=20:42:15 avg_step=1.33s recent_step=1.44s eta=16:19:38 +2026-08-26 00:38:02 | INFO | train rank=0 | step=55920/100000 progress=55.9% loss=0.041840 lr=4.435718e-06 elapsed=20:42:32 avg_step=1.33s recent_step=1.65s eta=16:19:27 +2026-08-26 00:38:14 | INFO | train rank=0 | step=55930/100000 progress=55.9% loss=0.065979 lr=4.434075e-06 elapsed=20:42:44 avg_step=1.33s recent_step=1.22s eta=16:19:13 +2026-08-26 00:38:29 | INFO | train rank=0 | step=55940/100000 progress=55.9% loss=0.055069 lr=4.432432e-06 elapsed=20:42:59 avg_step=1.33s recent_step=1.48s eta=16:19:00 +2026-08-26 00:38:41 | INFO | train rank=0 | step=55950/100000 progress=56.0% loss=0.072052 lr=4.430790e-06 elapsed=20:43:10 avg_step=1.33s recent_step=1.16s eta=16:18:46 +2026-08-26 00:38:52 | INFO | train rank=0 | step=55960/100000 progress=56.0% loss=0.038040 lr=4.429147e-06 elapsed=20:43:22 avg_step=1.33s recent_step=1.13s eta=16:18:31 +2026-08-26 00:39:06 | INFO | train rank=0 | step=55970/100000 progress=56.0% loss=0.047958 lr=4.427504e-06 elapsed=20:43:36 avg_step=1.33s recent_step=1.39s eta=16:18:18 +2026-08-26 00:39:20 | INFO | train rank=0 | step=55980/100000 progress=56.0% loss=0.041245 lr=4.425862e-06 elapsed=20:43:50 avg_step=1.33s recent_step=1.41s eta=16:18:05 +2026-08-26 00:39:32 | INFO | train rank=0 | step=55990/100000 progress=56.0% loss=0.047562 lr=4.424219e-06 elapsed=20:44:02 avg_step=1.33s recent_step=1.25s eta=16:17:51 +2026-08-26 00:39:50 | INFO | train rank=0 | step=56000/100000 progress=56.0% loss=0.034622 lr=4.422577e-06 elapsed=20:44:20 avg_step=1.33s recent_step=1.74s eta=16:17:41 +2026-08-26 00:40:01 | INFO | train rank=0 | step=56010/100000 progress=56.0% loss=0.050125 lr=4.420934e-06 elapsed=20:44:31 avg_step=1.33s recent_step=1.16s eta=16:17:26 +2026-08-26 00:40:14 | INFO | train rank=0 | step=56020/100000 progress=56.0% loss=0.027809 lr=4.419292e-06 elapsed=20:44:44 avg_step=1.33s recent_step=1.31s eta=16:17:13 +2026-08-26 00:40:30 | INFO | train rank=0 | step=56030/100000 progress=56.0% loss=0.032516 lr=4.417650e-06 elapsed=20:45:00 avg_step=1.33s recent_step=1.54s eta=16:17:01 +2026-08-26 00:40:46 | INFO | train rank=0 | step=56040/100000 progress=56.0% loss=0.039948 lr=4.416008e-06 elapsed=20:45:16 avg_step=1.33s recent_step=1.63s eta=16:16:50 +2026-08-26 00:40:58 | INFO | train rank=0 | step=56050/100000 progress=56.0% loss=0.040161 lr=4.414366e-06 elapsed=20:45:28 avg_step=1.33s recent_step=1.18s eta=16:16:36 +2026-08-26 00:41:12 | INFO | train rank=0 | step=56060/100000 progress=56.1% loss=0.049500 lr=4.412724e-06 elapsed=20:45:42 avg_step=1.33s recent_step=1.39s eta=16:16:23 +2026-08-26 00:41:29 | INFO | train rank=0 | step=56070/100000 progress=56.1% loss=0.057709 lr=4.411082e-06 elapsed=20:45:58 avg_step=1.33s recent_step=1.68s eta=16:16:12 +2026-08-26 00:41:42 | INFO | train rank=0 | step=56080/100000 progress=56.1% loss=0.049103 lr=4.409440e-06 elapsed=20:46:12 avg_step=1.33s recent_step=1.39s eta=16:15:59 +2026-08-26 00:41:56 | INFO | train rank=0 | step=56090/100000 progress=56.1% loss=0.066956 lr=4.407798e-06 elapsed=20:46:25 avg_step=1.33s recent_step=1.31s eta=16:15:46 +2026-08-26 00:42:07 | INFO | train rank=0 | step=56100/100000 progress=56.1% loss=0.029892 lr=4.406156e-06 elapsed=20:46:37 avg_step=1.33s recent_step=1.13s eta=16:15:31 +2026-08-26 00:42:25 | INFO | train rank=0 | step=56110/100000 progress=56.1% loss=0.056793 lr=4.404514e-06 elapsed=20:46:55 avg_step=1.33s recent_step=1.84s eta=16:15:21 +2026-08-26 00:42:37 | INFO | train rank=0 | step=56120/100000 progress=56.1% loss=0.048386 lr=4.402873e-06 elapsed=20:47:07 avg_step=1.33s recent_step=1.16s eta=16:15:07 +2026-08-26 00:42:51 | INFO | train rank=0 | step=56130/100000 progress=56.1% loss=0.087219 lr=4.401231e-06 elapsed=20:47:21 avg_step=1.33s recent_step=1.43s eta=16:14:54 +2026-08-26 00:43:03 | INFO | train rank=0 | step=56140/100000 progress=56.1% loss=0.043869 lr=4.399589e-06 elapsed=20:47:33 avg_step=1.33s recent_step=1.17s eta=16:14:40 +2026-08-26 00:43:16 | INFO | train rank=0 | step=56150/100000 progress=56.1% loss=0.040871 lr=4.397948e-06 elapsed=20:47:46 avg_step=1.33s recent_step=1.35s eta=16:14:26 +2026-08-26 00:43:35 | INFO | train rank=0 | step=56160/100000 progress=56.2% loss=0.053741 lr=4.396307e-06 elapsed=20:48:04 avg_step=1.33s recent_step=1.81s eta=16:14:17 +2026-08-26 00:43:46 | INFO | train rank=0 | step=56170/100000 progress=56.2% loss=0.083649 lr=4.394665e-06 elapsed=20:48:16 avg_step=1.33s recent_step=1.12s eta=16:14:02 +2026-08-26 00:43:57 | INFO | train rank=0 | step=56180/100000 progress=56.2% loss=0.047028 lr=4.393024e-06 elapsed=20:48:27 avg_step=1.33s recent_step=1.14s eta=16:13:47 +2026-08-26 00:44:09 | INFO | train rank=0 | step=56190/100000 progress=56.2% loss=0.031487 lr=4.391383e-06 elapsed=20:48:39 avg_step=1.33s recent_step=1.16s eta=16:13:32 +2026-08-26 00:44:22 | INFO | train rank=0 | step=56200/100000 progress=56.2% loss=0.051819 lr=4.389742e-06 elapsed=20:48:52 avg_step=1.33s recent_step=1.29s eta=16:13:18 +2026-08-26 00:44:33 | INFO | train rank=0 | step=56210/100000 progress=56.2% loss=0.056885 lr=4.388100e-06 elapsed=20:49:03 avg_step=1.33s recent_step=1.10s eta=16:13:03 +2026-08-26 00:44:44 | INFO | train rank=0 | step=56220/100000 progress=56.2% loss=0.061638 lr=4.386459e-06 elapsed=20:49:14 avg_step=1.33s recent_step=1.16s eta=16:12:49 +2026-08-26 00:44:56 | INFO | train rank=0 | step=56230/100000 progress=56.2% loss=0.060509 lr=4.384819e-06 elapsed=20:49:26 avg_step=1.33s recent_step=1.17s eta=16:12:34 +2026-08-26 00:45:11 | INFO | train rank=0 | step=56240/100000 progress=56.2% loss=0.040138 lr=4.383178e-06 elapsed=20:49:41 avg_step=1.33s recent_step=1.53s eta=16:12:22 +2026-08-26 00:45:24 | INFO | train rank=0 | step=56250/100000 progress=56.2% loss=0.041061 lr=4.381537e-06 elapsed=20:49:54 avg_step=1.33s recent_step=1.24s eta=16:12:08 +2026-08-26 00:45:37 | INFO | train rank=0 | step=56260/100000 progress=56.3% loss=0.047211 lr=4.379896e-06 elapsed=20:50:07 avg_step=1.33s recent_step=1.35s eta=16:11:55 +2026-08-26 00:45:49 | INFO | train rank=0 | step=56270/100000 progress=56.3% loss=0.043945 lr=4.378255e-06 elapsed=20:50:19 avg_step=1.33s recent_step=1.20s eta=16:11:41 +2026-08-26 00:46:05 | INFO | train rank=0 | step=56280/100000 progress=56.3% loss=0.038216 lr=4.376615e-06 elapsed=20:50:35 avg_step=1.33s recent_step=1.61s eta=16:11:30 +2026-08-26 00:46:21 | INFO | train rank=0 | step=56290/100000 progress=56.3% loss=0.036545 lr=4.374974e-06 elapsed=20:50:50 avg_step=1.33s recent_step=1.52s eta=16:11:18 +2026-08-26 00:46:32 | INFO | train rank=0 | step=56300/100000 progress=56.3% loss=0.059814 lr=4.373334e-06 elapsed=20:51:02 avg_step=1.33s recent_step=1.19s eta=16:11:03 +2026-08-26 00:46:46 | INFO | train rank=0 | step=56310/100000 progress=56.3% loss=0.045044 lr=4.371693e-06 elapsed=20:51:16 avg_step=1.33s recent_step=1.37s eta=16:10:50 +2026-08-26 00:47:00 | INFO | train rank=0 | step=56320/100000 progress=56.3% loss=0.036667 lr=4.370053e-06 elapsed=20:51:30 avg_step=1.33s recent_step=1.38s eta=16:10:37 +2026-08-26 00:47:12 | INFO | train rank=0 | step=56330/100000 progress=56.3% loss=0.049927 lr=4.368413e-06 elapsed=20:51:42 avg_step=1.33s recent_step=1.19s eta=16:10:23 +2026-08-26 00:47:26 | INFO | train rank=0 | step=56340/100000 progress=56.3% loss=0.038513 lr=4.366773e-06 elapsed=20:51:56 avg_step=1.33s recent_step=1.41s eta=16:10:10 +2026-08-26 00:47:43 | INFO | train rank=0 | step=56350/100000 progress=56.4% loss=0.040558 lr=4.365133e-06 elapsed=20:52:13 avg_step=1.33s recent_step=1.73s eta=16:10:00 +2026-08-26 00:47:55 | INFO | train rank=0 | step=56360/100000 progress=56.4% loss=0.050095 lr=4.363492e-06 elapsed=20:52:25 avg_step=1.33s recent_step=1.22s eta=16:09:45 +2026-08-26 00:48:11 | INFO | train rank=0 | step=56370/100000 progress=56.4% loss=0.053368 lr=4.361852e-06 elapsed=20:52:41 avg_step=1.33s recent_step=1.61s eta=16:09:34 +2026-08-26 00:48:23 | INFO | train rank=0 | step=56380/100000 progress=56.4% loss=0.031036 lr=4.360213e-06 elapsed=20:52:53 avg_step=1.33s recent_step=1.17s eta=16:09:20 +2026-08-26 00:48:38 | INFO | train rank=0 | step=56390/100000 progress=56.4% loss=0.052612 lr=4.358573e-06 elapsed=20:53:08 avg_step=1.33s recent_step=1.52s eta=16:09:08 +2026-08-26 00:48:51 | INFO | train rank=0 | step=56400/100000 progress=56.4% loss=0.036758 lr=4.356933e-06 elapsed=20:53:21 avg_step=1.33s recent_step=1.23s eta=16:08:54 +2026-08-26 00:49:03 | INFO | train rank=0 | step=56410/100000 progress=56.4% loss=0.107056 lr=4.355293e-06 elapsed=20:53:33 avg_step=1.33s recent_step=1.25s eta=16:08:40 +2026-08-26 00:49:15 | INFO | train rank=0 | step=56420/100000 progress=56.4% loss=0.042068 lr=4.353654e-06 elapsed=20:53:44 avg_step=1.33s recent_step=1.14s eta=16:08:25 +2026-08-26 00:49:29 | INFO | train rank=0 | step=56430/100000 progress=56.4% loss=0.047821 lr=4.352014e-06 elapsed=20:53:59 avg_step=1.33s recent_step=1.48s eta=16:08:13 +2026-08-26 00:49:42 | INFO | train rank=0 | step=56440/100000 progress=56.4% loss=0.059349 lr=4.350375e-06 elapsed=20:54:12 avg_step=1.33s recent_step=1.26s eta=16:07:59 +2026-08-26 00:49:56 | INFO | train rank=0 | step=56450/100000 progress=56.5% loss=0.040695 lr=4.348735e-06 elapsed=20:54:26 avg_step=1.33s recent_step=1.41s eta=16:07:46 +2026-08-26 00:50:09 | INFO | train rank=0 | step=56460/100000 progress=56.5% loss=0.038342 lr=4.347096e-06 elapsed=20:54:39 avg_step=1.33s recent_step=1.30s eta=16:07:32 +2026-08-26 00:50:27 | INFO | train rank=0 | step=56470/100000 progress=56.5% loss=0.049454 lr=4.345457e-06 elapsed=20:54:57 avg_step=1.33s recent_step=1.75s eta=16:07:22 +2026-08-26 00:50:40 | INFO | train rank=0 | step=56480/100000 progress=56.5% loss=0.086670 lr=4.343817e-06 elapsed=20:55:10 avg_step=1.33s recent_step=1.36s eta=16:07:09 +2026-08-26 00:50:53 | INFO | train rank=0 | step=56490/100000 progress=56.5% loss=0.045212 lr=4.342178e-06 elapsed=20:55:23 avg_step=1.33s recent_step=1.26s eta=16:06:55 +2026-08-26 00:51:06 | INFO | train rank=0 | step=56500/100000 progress=56.5% loss=0.038795 lr=4.340539e-06 elapsed=20:55:36 avg_step=1.33s recent_step=1.29s eta=16:06:42 +2026-08-26 00:51:23 | INFO | train rank=0 | step=56510/100000 progress=56.5% loss=0.034302 lr=4.338900e-06 elapsed=20:55:53 avg_step=1.33s recent_step=1.70s eta=16:06:31 +2026-08-26 00:51:41 | INFO | train rank=0 | step=56520/100000 progress=56.5% loss=0.054413 lr=4.337261e-06 elapsed=20:56:11 avg_step=1.33s recent_step=1.83s eta=16:06:22 +2026-08-26 00:51:52 | INFO | train rank=0 | step=56530/100000 progress=56.5% loss=0.044403 lr=4.335622e-06 elapsed=20:56:22 avg_step=1.33s recent_step=1.10s eta=16:06:07 +2026-08-26 00:52:05 | INFO | train rank=0 | step=56540/100000 progress=56.5% loss=0.082153 lr=4.333984e-06 elapsed=20:56:35 avg_step=1.33s recent_step=1.33s eta=16:05:53 +2026-08-26 00:52:21 | INFO | train rank=0 | step=56550/100000 progress=56.5% loss=0.031250 lr=4.332345e-06 elapsed=20:56:51 avg_step=1.33s recent_step=1.59s eta=16:05:42 +2026-08-26 00:52:32 | INFO | train rank=0 | step=56560/100000 progress=56.6% loss=0.044334 lr=4.330706e-06 elapsed=20:57:02 avg_step=1.33s recent_step=1.12s eta=16:05:27 +2026-08-26 00:52:44 | INFO | train rank=0 | step=56570/100000 progress=56.6% loss=0.066711 lr=4.329068e-06 elapsed=20:57:14 avg_step=1.33s recent_step=1.20s eta=16:05:12 +2026-08-26 00:53:01 | INFO | train rank=0 | step=56580/100000 progress=56.6% loss=0.056618 lr=4.327429e-06 elapsed=20:57:31 avg_step=1.33s recent_step=1.64s eta=16:05:01 +2026-08-26 00:53:15 | INFO | train rank=0 | step=56590/100000 progress=56.6% loss=0.034546 lr=4.325791e-06 elapsed=20:57:45 avg_step=1.33s recent_step=1.43s eta=16:04:49 +2026-08-26 00:53:27 | INFO | train rank=0 | step=56600/100000 progress=56.6% loss=0.044495 lr=4.324153e-06 elapsed=20:57:57 avg_step=1.33s recent_step=1.16s eta=16:04:34 +2026-08-26 00:53:38 | INFO | train rank=0 | step=56610/100000 progress=56.6% loss=0.033325 lr=4.322514e-06 elapsed=20:58:08 avg_step=1.33s recent_step=1.17s eta=16:04:20 +2026-08-26 00:53:50 | INFO | train rank=0 | step=56620/100000 progress=56.6% loss=0.042435 lr=4.320876e-06 elapsed=20:58:20 avg_step=1.33s recent_step=1.21s eta=16:04:05 +2026-08-26 00:54:04 | INFO | train rank=0 | step=56630/100000 progress=56.6% loss=0.047226 lr=4.319238e-06 elapsed=20:58:34 avg_step=1.33s recent_step=1.32s eta=16:03:52 +2026-08-26 00:54:19 | INFO | train rank=0 | step=56640/100000 progress=56.6% loss=0.039948 lr=4.317600e-06 elapsed=20:58:49 avg_step=1.33s recent_step=1.58s eta=16:03:40 +2026-08-26 00:54:32 | INFO | train rank=0 | step=56650/100000 progress=56.6% loss=0.029816 lr=4.315962e-06 elapsed=20:59:02 avg_step=1.33s recent_step=1.26s eta=16:03:26 +2026-08-26 00:54:48 | INFO | train rank=0 | step=56660/100000 progress=56.7% loss=0.087616 lr=4.314324e-06 elapsed=20:59:18 avg_step=1.33s recent_step=1.59s eta=16:03:15 +2026-08-26 00:55:01 | INFO | train rank=0 | step=56670/100000 progress=56.7% loss=0.042450 lr=4.312686e-06 elapsed=20:59:31 avg_step=1.33s recent_step=1.32s eta=16:03:02 +2026-08-26 00:55:13 | INFO | train rank=0 | step=56680/100000 progress=56.7% loss=0.120605 lr=4.311049e-06 elapsed=20:59:43 avg_step=1.33s recent_step=1.16s eta=16:02:47 +2026-08-26 00:55:24 | INFO | train rank=0 | step=56690/100000 progress=56.7% loss=0.045494 lr=4.309411e-06 elapsed=20:59:54 avg_step=1.33s recent_step=1.14s eta=16:02:32 +2026-08-26 00:55:40 | INFO | train rank=0 | step=56700/100000 progress=56.7% loss=0.042969 lr=4.307773e-06 elapsed=21:00:10 avg_step=1.33s recent_step=1.59s eta=16:02:21 +2026-08-26 00:55:51 | INFO | train rank=0 | step=56710/100000 progress=56.7% loss=0.069855 lr=4.306136e-06 elapsed=21:00:21 avg_step=1.33s recent_step=1.10s eta=16:02:06 +2026-08-26 00:56:09 | INFO | train rank=0 | step=56720/100000 progress=56.7% loss=0.037216 lr=4.304498e-06 elapsed=21:00:39 avg_step=1.33s recent_step=1.84s eta=16:01:56 +2026-08-26 00:56:21 | INFO | train rank=0 | step=56730/100000 progress=56.7% loss=0.060364 lr=4.302861e-06 elapsed=21:00:51 avg_step=1.33s recent_step=1.18s eta=16:01:42 +2026-08-26 00:56:35 | INFO | train rank=0 | step=56740/100000 progress=56.7% loss=0.045258 lr=4.301224e-06 elapsed=21:01:05 avg_step=1.33s recent_step=1.40s eta=16:01:29 +2026-08-26 00:56:48 | INFO | train rank=0 | step=56750/100000 progress=56.8% loss=0.054428 lr=4.299587e-06 elapsed=21:01:18 avg_step=1.33s recent_step=1.26s eta=16:01:15 +2026-08-26 00:57:02 | INFO | train rank=0 | step=56760/100000 progress=56.8% loss=0.041031 lr=4.297949e-06 elapsed=21:01:32 avg_step=1.33s recent_step=1.46s eta=16:01:03 +2026-08-26 00:57:15 | INFO | train rank=0 | step=56770/100000 progress=56.8% loss=0.059067 lr=4.296312e-06 elapsed=21:01:45 avg_step=1.33s recent_step=1.29s eta=16:00:49 +2026-08-26 00:57:30 | INFO | train rank=0 | step=56780/100000 progress=56.8% loss=0.049805 lr=4.294675e-06 elapsed=21:02:00 avg_step=1.33s recent_step=1.48s eta=16:00:37 +2026-08-26 00:57:42 | INFO | train rank=0 | step=56790/100000 progress=56.8% loss=0.058136 lr=4.293038e-06 elapsed=21:02:12 avg_step=1.33s recent_step=1.15s eta=16:00:22 +2026-08-26 00:57:55 | INFO | train rank=0 | step=56800/100000 progress=56.8% loss=0.029472 lr=4.291402e-06 elapsed=21:02:25 avg_step=1.33s recent_step=1.29s eta=16:00:08 +2026-08-26 00:58:06 | INFO | train rank=0 | step=56810/100000 progress=56.8% loss=0.042206 lr=4.289765e-06 elapsed=21:02:36 avg_step=1.33s recent_step=1.16s eta=15:59:54 +2026-08-26 00:58:19 | INFO | train rank=0 | step=56820/100000 progress=56.8% loss=0.043182 lr=4.288128e-06 elapsed=21:02:48 avg_step=1.33s recent_step=1.23s eta=15:59:40 +2026-08-26 00:58:30 | INFO | train rank=0 | step=56830/100000 progress=56.8% loss=0.054474 lr=4.286492e-06 elapsed=21:03:00 avg_step=1.33s recent_step=1.14s eta=15:59:25 +2026-08-26 00:58:45 | INFO | train rank=0 | step=56840/100000 progress=56.8% loss=0.045578 lr=4.284855e-06 elapsed=21:03:14 avg_step=1.33s recent_step=1.45s eta=15:59:12 +2026-08-26 00:58:59 | INFO | train rank=0 | step=56850/100000 progress=56.9% loss=0.044479 lr=4.283219e-06 elapsed=21:03:29 avg_step=1.33s recent_step=1.43s eta=15:59:00 +2026-08-26 00:59:11 | INFO | train rank=0 | step=56860/100000 progress=56.9% loss=0.053467 lr=4.281582e-06 elapsed=21:03:41 avg_step=1.33s recent_step=1.21s eta=15:58:46 +2026-08-26 00:59:23 | INFO | train rank=0 | step=56870/100000 progress=56.9% loss=0.111938 lr=4.279946e-06 elapsed=21:03:53 avg_step=1.33s recent_step=1.21s eta=15:58:31 +2026-08-26 00:59:37 | INFO | train rank=0 | step=56880/100000 progress=56.9% loss=0.031998 lr=4.278310e-06 elapsed=21:04:07 avg_step=1.33s recent_step=1.41s eta=15:58:18 +2026-08-26 00:59:51 | INFO | train rank=0 | step=56890/100000 progress=56.9% loss=0.064041 lr=4.276674e-06 elapsed=21:04:20 avg_step=1.33s recent_step=1.35s eta=15:58:05 +2026-08-26 01:00:03 | INFO | train rank=0 | step=56900/100000 progress=56.9% loss=0.029915 lr=4.275038e-06 elapsed=21:04:33 avg_step=1.33s recent_step=1.27s eta=15:57:51 +2026-08-26 01:00:15 | INFO | train rank=0 | step=56910/100000 progress=56.9% loss=0.064423 lr=4.273402e-06 elapsed=21:04:45 avg_step=1.33s recent_step=1.15s eta=15:57:37 +2026-08-26 01:00:27 | INFO | train rank=0 | step=56920/100000 progress=56.9% loss=0.036896 lr=4.271766e-06 elapsed=21:04:57 avg_step=1.33s recent_step=1.24s eta=15:57:23 +2026-08-26 01:00:38 | INFO | train rank=0 | step=56930/100000 progress=56.9% loss=0.059967 lr=4.270130e-06 elapsed=21:05:08 avg_step=1.33s recent_step=1.12s eta=15:57:08 +2026-08-26 01:00:58 | INFO | train rank=0 | step=56940/100000 progress=56.9% loss=0.065155 lr=4.268494e-06 elapsed=21:05:28 avg_step=1.33s recent_step=1.99s eta=15:56:59 +2026-08-26 01:01:13 | INFO | train rank=0 | step=56950/100000 progress=57.0% loss=0.055313 lr=4.266859e-06 elapsed=21:05:43 avg_step=1.33s recent_step=1.50s eta=15:56:47 +2026-08-26 01:01:28 | INFO | train rank=0 | step=56960/100000 progress=57.0% loss=0.056320 lr=4.265223e-06 elapsed=21:05:57 avg_step=1.33s recent_step=1.43s eta=15:56:35 +2026-08-26 01:01:43 | INFO | train rank=0 | step=56970/100000 progress=57.0% loss=0.035095 lr=4.263588e-06 elapsed=21:06:13 avg_step=1.33s recent_step=1.58s eta=15:56:23 +2026-08-26 01:01:54 | INFO | train rank=0 | step=56980/100000 progress=57.0% loss=0.053482 lr=4.261952e-06 elapsed=21:06:24 avg_step=1.33s recent_step=1.10s eta=15:56:08 +2026-08-26 01:02:07 | INFO | train rank=0 | step=56990/100000 progress=57.0% loss=0.038986 lr=4.260317e-06 elapsed=21:06:37 avg_step=1.33s recent_step=1.23s eta=15:55:54 +2026-08-26 01:02:18 | INFO | train rank=0 | step=57000/100000 progress=57.0% loss=0.059479 lr=4.258682e-06 elapsed=21:06:48 avg_step=1.33s recent_step=1.16s eta=15:55:39 +2026-08-26 01:02:31 | INFO | train rank=0 | step=57010/100000 progress=57.0% loss=0.053375 lr=4.257047e-06 elapsed=21:07:00 avg_step=1.33s recent_step=1.24s eta=15:55:25 +2026-08-26 01:02:47 | INFO | train rank=0 | step=57020/100000 progress=57.0% loss=0.132568 lr=4.255412e-06 elapsed=21:07:17 avg_step=1.33s recent_step=1.64s eta=15:55:14 +2026-08-26 01:03:01 | INFO | train rank=0 | step=57030/100000 progress=57.0% loss=0.053558 lr=4.253777e-06 elapsed=21:07:31 avg_step=1.33s recent_step=1.40s eta=15:55:01 +2026-08-26 01:03:13 | INFO | train rank=0 | step=57040/100000 progress=57.0% loss=0.042191 lr=4.252142e-06 elapsed=21:07:42 avg_step=1.33s recent_step=1.16s eta=15:54:47 +2026-08-26 01:03:26 | INFO | train rank=0 | step=57050/100000 progress=57.0% loss=0.051224 lr=4.250507e-06 elapsed=21:07:56 avg_step=1.33s recent_step=1.38s eta=15:54:34 +2026-08-26 01:03:40 | INFO | train rank=0 | step=57060/100000 progress=57.1% loss=0.051239 lr=4.248872e-06 elapsed=21:08:10 avg_step=1.33s recent_step=1.41s eta=15:54:21 +2026-08-26 01:03:55 | INFO | train rank=0 | step=57070/100000 progress=57.1% loss=0.036865 lr=4.247237e-06 elapsed=21:08:24 avg_step=1.33s recent_step=1.41s eta=15:54:08 +2026-08-26 01:04:12 | INFO | train rank=0 | step=57080/100000 progress=57.1% loss=0.035751 lr=4.245603e-06 elapsed=21:08:42 avg_step=1.33s recent_step=1.72s eta=15:53:58 +2026-08-26 01:04:26 | INFO | train rank=0 | step=57090/100000 progress=57.1% loss=0.052887 lr=4.243968e-06 elapsed=21:08:56 avg_step=1.33s recent_step=1.41s eta=15:53:45 +2026-08-26 01:04:38 | INFO | train rank=0 | step=57100/100000 progress=57.1% loss=0.042374 lr=4.242334e-06 elapsed=21:09:08 avg_step=1.33s recent_step=1.20s eta=15:53:31 +2026-08-26 01:04:50 | INFO | train rank=0 | step=57110/100000 progress=57.1% loss=0.054276 lr=4.240700e-06 elapsed=21:09:20 avg_step=1.33s recent_step=1.20s eta=15:53:16 +2026-08-26 01:05:04 | INFO | train rank=0 | step=57120/100000 progress=57.1% loss=0.043869 lr=4.239065e-06 elapsed=21:09:34 avg_step=1.33s recent_step=1.44s eta=15:53:04 +2026-08-26 01:05:23 | INFO | train rank=0 | step=57130/100000 progress=57.1% loss=0.045998 lr=4.237431e-06 elapsed=21:09:53 avg_step=1.33s recent_step=1.85s eta=15:52:54 +2026-08-26 01:05:34 | INFO | train rank=0 | step=57140/100000 progress=57.1% loss=0.051697 lr=4.235797e-06 elapsed=21:10:04 avg_step=1.33s recent_step=1.11s eta=15:52:39 +2026-08-26 01:05:46 | INFO | train rank=0 | step=57150/100000 progress=57.1% loss=0.047180 lr=4.234163e-06 elapsed=21:10:16 avg_step=1.33s recent_step=1.19s eta=15:52:25 +2026-08-26 01:05:57 | INFO | train rank=0 | step=57160/100000 progress=57.2% loss=0.031738 lr=4.232529e-06 elapsed=21:10:27 avg_step=1.33s recent_step=1.15s eta=15:52:10 +2026-08-26 01:06:09 | INFO | train rank=0 | step=57170/100000 progress=57.2% loss=0.036911 lr=4.230895e-06 elapsed=21:10:39 avg_step=1.33s recent_step=1.15s eta=15:51:56 +2026-08-26 01:06:20 | INFO | train rank=0 | step=57180/100000 progress=57.2% loss=0.048874 lr=4.229262e-06 elapsed=21:10:50 avg_step=1.33s recent_step=1.16s eta=15:51:41 +2026-08-26 01:06:35 | INFO | train rank=0 | step=57190/100000 progress=57.2% loss=0.037704 lr=4.227628e-06 elapsed=21:11:05 avg_step=1.33s recent_step=1.47s eta=15:51:29 +2026-08-26 01:06:53 | INFO | train rank=0 | step=57200/100000 progress=57.2% loss=0.040085 lr=4.225994e-06 elapsed=21:11:23 avg_step=1.33s recent_step=1.79s eta=15:51:19 +2026-08-26 01:07:08 | INFO | train rank=0 | step=57210/100000 progress=57.2% loss=0.032158 lr=4.224361e-06 elapsed=21:11:37 avg_step=1.33s recent_step=1.45s eta=15:51:06 +2026-08-26 01:07:19 | INFO | train rank=0 | step=57220/100000 progress=57.2% loss=0.077271 lr=4.222727e-06 elapsed=21:11:49 avg_step=1.33s recent_step=1.16s eta=15:50:52 +2026-08-26 01:07:31 | INFO | train rank=0 | step=57230/100000 progress=57.2% loss=0.047394 lr=4.221094e-06 elapsed=21:12:00 avg_step=1.33s recent_step=1.14s eta=15:50:37 +2026-08-26 01:07:42 | INFO | train rank=0 | step=57240/100000 progress=57.2% loss=0.067596 lr=4.219461e-06 elapsed=21:12:12 avg_step=1.33s recent_step=1.17s eta=15:50:22 +2026-08-26 01:07:56 | INFO | train rank=0 | step=57250/100000 progress=57.2% loss=0.068054 lr=4.217828e-06 elapsed=21:12:26 avg_step=1.33s recent_step=1.41s eta=15:50:10 +2026-08-26 01:08:08 | INFO | train rank=0 | step=57260/100000 progress=57.3% loss=0.041504 lr=4.216195e-06 elapsed=21:12:38 avg_step=1.33s recent_step=1.12s eta=15:49:55 +2026-08-26 01:08:23 | INFO | train rank=0 | step=57270/100000 progress=57.3% loss=0.050323 lr=4.214562e-06 elapsed=21:12:53 avg_step=1.33s recent_step=1.58s eta=15:49:43 +2026-08-26 01:08:40 | INFO | train rank=0 | step=57280/100000 progress=57.3% loss=0.037430 lr=4.212929e-06 elapsed=21:13:10 avg_step=1.33s recent_step=1.65s eta=15:49:32 +2026-08-26 01:08:51 | INFO | train rank=0 | step=57290/100000 progress=57.3% loss=0.044815 lr=4.211296e-06 elapsed=21:13:21 avg_step=1.33s recent_step=1.08s eta=15:49:17 +2026-08-26 01:09:02 | INFO | train rank=0 | step=57300/100000 progress=57.3% loss=0.035782 lr=4.209663e-06 elapsed=21:13:32 avg_step=1.33s recent_step=1.13s eta=15:49:02 +2026-08-26 01:09:19 | INFO | train rank=0 | step=57310/100000 progress=57.3% loss=0.046570 lr=4.208031e-06 elapsed=21:13:49 avg_step=1.33s recent_step=1.65s eta=15:48:51 +2026-08-26 01:09:35 | INFO | train rank=0 | step=57320/100000 progress=57.3% loss=0.037163 lr=4.206398e-06 elapsed=21:14:05 avg_step=1.33s recent_step=1.60s eta=15:48:40 +2026-08-26 01:09:47 | INFO | train rank=0 | step=57330/100000 progress=57.3% loss=0.051239 lr=4.204766e-06 elapsed=21:14:17 avg_step=1.33s recent_step=1.19s eta=15:48:26 +2026-08-26 01:10:01 | INFO | train rank=0 | step=57340/100000 progress=57.3% loss=0.042877 lr=4.203133e-06 elapsed=21:14:31 avg_step=1.33s recent_step=1.46s eta=15:48:13 +2026-08-26 01:10:13 | INFO | train rank=0 | step=57350/100000 progress=57.4% loss=0.034027 lr=4.201501e-06 elapsed=21:14:43 avg_step=1.33s recent_step=1.14s eta=15:47:58 +2026-08-26 01:10:26 | INFO | train rank=0 | step=57360/100000 progress=57.4% loss=0.095215 lr=4.199869e-06 elapsed=21:14:56 avg_step=1.33s recent_step=1.34s eta=15:47:45 +2026-08-26 01:10:37 | INFO | train rank=0 | step=57370/100000 progress=57.4% loss=0.069580 lr=4.198237e-06 elapsed=21:15:07 avg_step=1.33s recent_step=1.09s eta=15:47:30 +2026-08-26 01:10:55 | INFO | train rank=0 | step=57380/100000 progress=57.4% loss=0.049927 lr=4.196605e-06 elapsed=21:15:24 avg_step=1.33s recent_step=1.75s eta=15:47:20 +2026-08-26 01:11:06 | INFO | train rank=0 | step=57390/100000 progress=57.4% loss=0.067902 lr=4.194973e-06 elapsed=21:15:36 avg_step=1.33s recent_step=1.19s eta=15:47:05 +2026-08-26 01:11:18 | INFO | train rank=0 | step=57400/100000 progress=57.4% loss=0.049690 lr=4.193341e-06 elapsed=21:15:48 avg_step=1.33s recent_step=1.18s eta=15:46:51 +2026-08-26 01:11:30 | INFO | train rank=0 | step=57410/100000 progress=57.4% loss=0.073059 lr=4.191709e-06 elapsed=21:16:00 avg_step=1.33s recent_step=1.20s eta=15:46:37 +2026-08-26 01:11:46 | INFO | train rank=0 | step=57420/100000 progress=57.4% loss=0.061829 lr=4.190077e-06 elapsed=21:16:16 avg_step=1.33s recent_step=1.54s eta=15:46:25 +2026-08-26 01:11:58 | INFO | train rank=0 | step=57430/100000 progress=57.4% loss=0.046814 lr=4.188446e-06 elapsed=21:16:27 avg_step=1.33s recent_step=1.19s eta=15:46:10 +2026-08-26 01:12:14 | INFO | train rank=0 | step=57440/100000 progress=57.4% loss=0.074249 lr=4.186814e-06 elapsed=21:16:44 avg_step=1.33s recent_step=1.63s eta=15:45:59 +2026-08-26 01:12:25 | INFO | train rank=0 | step=57450/100000 progress=57.5% loss=0.058014 lr=4.185183e-06 elapsed=21:16:55 avg_step=1.33s recent_step=1.10s eta=15:45:44 +2026-08-26 01:12:37 | INFO | train rank=0 | step=57460/100000 progress=57.5% loss=0.062256 lr=4.183551e-06 elapsed=21:17:07 avg_step=1.33s recent_step=1.18s eta=15:45:30 +2026-08-26 01:12:48 | INFO | train rank=0 | step=57470/100000 progress=57.5% loss=0.064484 lr=4.181920e-06 elapsed=21:17:18 avg_step=1.33s recent_step=1.16s eta=15:45:15 +2026-08-26 01:13:09 | INFO | train rank=0 | step=57480/100000 progress=57.5% loss=0.041443 lr=4.180289e-06 elapsed=21:17:39 avg_step=1.33s recent_step=2.10s eta=15:45:07 +2026-08-26 01:13:21 | INFO | train rank=0 | step=57490/100000 progress=57.5% loss=0.061844 lr=4.178658e-06 elapsed=21:17:51 avg_step=1.33s recent_step=1.16s eta=15:44:53 +2026-08-26 01:13:32 | INFO | train rank=0 | step=57500/100000 progress=57.5% loss=0.045135 lr=4.177027e-06 elapsed=21:18:02 avg_step=1.33s recent_step=1.11s eta=15:44:38 +2026-08-26 01:13:44 | INFO | train rank=0 | step=57510/100000 progress=57.5% loss=0.046738 lr=4.175396e-06 elapsed=21:18:13 avg_step=1.33s recent_step=1.15s eta=15:44:23 +2026-08-26 01:14:00 | INFO | train rank=0 | step=57520/100000 progress=57.5% loss=0.044662 lr=4.173765e-06 elapsed=21:18:30 avg_step=1.33s recent_step=1.68s eta=15:44:12 +2026-08-26 01:14:12 | INFO | train rank=0 | step=57530/100000 progress=57.5% loss=0.077026 lr=4.172135e-06 elapsed=21:18:42 avg_step=1.33s recent_step=1.18s eta=15:43:58 +2026-08-26 01:14:24 | INFO | train rank=0 | step=57540/100000 progress=57.5% loss=0.061066 lr=4.170504e-06 elapsed=21:18:54 avg_step=1.33s recent_step=1.17s eta=15:43:43 +2026-08-26 01:14:37 | INFO | train rank=0 | step=57550/100000 progress=57.6% loss=0.070496 lr=4.168874e-06 elapsed=21:19:07 avg_step=1.33s recent_step=1.33s eta=15:43:30 +2026-08-26 01:14:49 | INFO | train rank=0 | step=57560/100000 progress=57.6% loss=0.064163 lr=4.167243e-06 elapsed=21:19:19 avg_step=1.33s recent_step=1.23s eta=15:43:16 +2026-08-26 01:15:02 | INFO | train rank=0 | step=57570/100000 progress=57.6% loss=0.036728 lr=4.165613e-06 elapsed=21:19:32 avg_step=1.33s recent_step=1.28s eta=15:43:02 +2026-08-26 01:15:14 | INFO | train rank=0 | step=57580/100000 progress=57.6% loss=0.061188 lr=4.163983e-06 elapsed=21:19:44 avg_step=1.33s recent_step=1.16s eta=15:42:48 +2026-08-26 01:15:30 | INFO | train rank=0 | step=57590/100000 progress=57.6% loss=0.032913 lr=4.162352e-06 elapsed=21:20:00 avg_step=1.33s recent_step=1.63s eta=15:42:36 +2026-08-26 01:15:42 | INFO | train rank=0 | step=57600/100000 progress=57.6% loss=0.072449 lr=4.160722e-06 elapsed=21:20:12 avg_step=1.33s recent_step=1.18s eta=15:42:22 +2026-08-26 01:15:53 | INFO | train rank=0 | step=57610/100000 progress=57.6% loss=0.035400 lr=4.159092e-06 elapsed=21:20:23 avg_step=1.33s recent_step=1.14s eta=15:42:07 +2026-08-26 01:16:10 | INFO | train rank=0 | step=57620/100000 progress=57.6% loss=0.037971 lr=4.157463e-06 elapsed=21:20:40 avg_step=1.33s recent_step=1.65s eta=15:41:56 +2026-08-26 01:16:21 | INFO | train rank=0 | step=57630/100000 progress=57.6% loss=0.062836 lr=4.155833e-06 elapsed=21:20:51 avg_step=1.33s recent_step=1.17s eta=15:41:42 +2026-08-26 01:16:36 | INFO | train rank=0 | step=57640/100000 progress=57.6% loss=0.039993 lr=4.154203e-06 elapsed=21:21:06 avg_step=1.33s recent_step=1.44s eta=15:41:29 +2026-08-26 01:16:51 | INFO | train rank=0 | step=57650/100000 progress=57.6% loss=0.060394 lr=4.152573e-06 elapsed=21:21:21 avg_step=1.33s recent_step=1.56s eta=15:41:17 +2026-08-26 01:17:03 | INFO | train rank=0 | step=57660/100000 progress=57.7% loss=0.055603 lr=4.150944e-06 elapsed=21:21:33 avg_step=1.33s recent_step=1.13s eta=15:41:02 +2026-08-26 01:17:14 | INFO | train rank=0 | step=57670/100000 progress=57.7% loss=0.050079 lr=4.149315e-06 elapsed=21:21:44 avg_step=1.33s recent_step=1.16s eta=15:40:48 +2026-08-26 01:17:29 | INFO | train rank=0 | step=57680/100000 progress=57.7% loss=0.036354 lr=4.147685e-06 elapsed=21:21:59 avg_step=1.33s recent_step=1.45s eta=15:40:35 +2026-08-26 01:17:40 | INFO | train rank=0 | step=57690/100000 progress=57.7% loss=0.042816 lr=4.146056e-06 elapsed=21:22:10 avg_step=1.33s recent_step=1.15s eta=15:40:21 +2026-08-26 01:17:52 | INFO | train rank=0 | step=57700/100000 progress=57.7% loss=0.068481 lr=4.144427e-06 elapsed=21:22:22 avg_step=1.33s recent_step=1.13s eta=15:40:06 +2026-08-26 01:18:08 | INFO | train rank=0 | step=57710/100000 progress=57.7% loss=0.049355 lr=4.142798e-06 elapsed=21:22:38 avg_step=1.33s recent_step=1.62s eta=15:39:55 +2026-08-26 01:18:24 | INFO | train rank=0 | step=57720/100000 progress=57.7% loss=0.043030 lr=4.141169e-06 elapsed=21:22:53 avg_step=1.33s recent_step=1.57s eta=15:39:43 +2026-08-26 01:18:36 | INFO | train rank=0 | step=57730/100000 progress=57.7% loss=0.050964 lr=4.139540e-06 elapsed=21:23:06 avg_step=1.33s recent_step=1.24s eta=15:39:29 +2026-08-26 01:18:47 | INFO | train rank=0 | step=57740/100000 progress=57.7% loss=0.053513 lr=4.137911e-06 elapsed=21:23:17 avg_step=1.33s recent_step=1.13s eta=15:39:14 +2026-08-26 01:18:59 | INFO | train rank=0 | step=57750/100000 progress=57.8% loss=0.038498 lr=4.136283e-06 elapsed=21:23:29 avg_step=1.33s recent_step=1.16s eta=15:39:00 +2026-08-26 01:19:14 | INFO | train rank=0 | step=57760/100000 progress=57.8% loss=0.028488 lr=4.134654e-06 elapsed=21:23:44 avg_step=1.33s recent_step=1.47s eta=15:38:47 +2026-08-26 01:19:30 | INFO | train rank=0 | step=57770/100000 progress=57.8% loss=0.039703 lr=4.133026e-06 elapsed=21:24:00 avg_step=1.33s recent_step=1.66s eta=15:38:36 +2026-08-26 01:19:42 | INFO | train rank=0 | step=57780/100000 progress=57.8% loss=0.030632 lr=4.131397e-06 elapsed=21:24:12 avg_step=1.33s recent_step=1.21s eta=15:38:22 +2026-08-26 01:19:56 | INFO | train rank=0 | step=57790/100000 progress=57.8% loss=0.032852 lr=4.129769e-06 elapsed=21:24:26 avg_step=1.33s recent_step=1.40s eta=15:38:09 +2026-08-26 01:20:08 | INFO | train rank=0 | step=57800/100000 progress=57.8% loss=0.054199 lr=4.128141e-06 elapsed=21:24:37 avg_step=1.33s recent_step=1.12s eta=15:37:54 +2026-08-26 01:20:22 | INFO | train rank=0 | step=57810/100000 progress=57.8% loss=0.033447 lr=4.126513e-06 elapsed=21:24:52 avg_step=1.33s recent_step=1.47s eta=15:37:42 +2026-08-26 01:20:35 | INFO | train rank=0 | step=57820/100000 progress=57.8% loss=0.048035 lr=4.124885e-06 elapsed=21:25:05 avg_step=1.33s recent_step=1.27s eta=15:37:28 +2026-08-26 01:20:50 | INFO | train rank=0 | step=57830/100000 progress=57.8% loss=0.044785 lr=4.123257e-06 elapsed=21:25:19 avg_step=1.33s recent_step=1.46s eta=15:37:16 +2026-08-26 01:21:04 | INFO | train rank=0 | step=57840/100000 progress=57.8% loss=0.050430 lr=4.121629e-06 elapsed=21:25:34 avg_step=1.33s recent_step=1.48s eta=15:37:04 +2026-08-26 01:21:16 | INFO | train rank=0 | step=57850/100000 progress=57.9% loss=0.029518 lr=4.120001e-06 elapsed=21:25:46 avg_step=1.33s recent_step=1.14s eta=15:36:49 +2026-08-26 01:21:28 | INFO | train rank=0 | step=57860/100000 progress=57.9% loss=0.083038 lr=4.118374e-06 elapsed=21:25:57 avg_step=1.33s recent_step=1.19s eta=15:36:34 +2026-08-26 01:21:41 | INFO | train rank=0 | step=57870/100000 progress=57.9% loss=0.041595 lr=4.116746e-06 elapsed=21:26:10 avg_step=1.33s recent_step=1.30s eta=15:36:21 +2026-08-26 01:22:00 | INFO | train rank=0 | step=57880/100000 progress=57.9% loss=0.042793 lr=4.115119e-06 elapsed=21:26:30 avg_step=1.33s recent_step=1.93s eta=15:36:12 +2026-08-26 01:22:12 | INFO | train rank=0 | step=57890/100000 progress=57.9% loss=0.043083 lr=4.113491e-06 elapsed=21:26:42 avg_step=1.33s recent_step=1.19s eta=15:35:58 +2026-08-26 01:22:23 | INFO | train rank=0 | step=57900/100000 progress=57.9% loss=0.048157 lr=4.111864e-06 elapsed=21:26:53 avg_step=1.33s recent_step=1.10s eta=15:35:42 +2026-08-26 01:22:41 | INFO | train rank=0 | step=57910/100000 progress=57.9% loss=0.042694 lr=4.110237e-06 elapsed=21:27:11 avg_step=1.33s recent_step=1.82s eta=15:35:33 +2026-08-26 01:22:54 | INFO | train rank=0 | step=57920/100000 progress=57.9% loss=0.037025 lr=4.108610e-06 elapsed=21:27:24 avg_step=1.33s recent_step=1.28s eta=15:35:19 +2026-08-26 01:23:05 | INFO | train rank=0 | step=57930/100000 progress=57.9% loss=0.039612 lr=4.106983e-06 elapsed=21:27:35 avg_step=1.33s recent_step=1.14s eta=15:35:04 +2026-08-26 01:23:20 | INFO | train rank=0 | step=57940/100000 progress=57.9% loss=0.057831 lr=4.105356e-06 elapsed=21:27:50 avg_step=1.33s recent_step=1.49s eta=15:34:52 +2026-08-26 01:23:34 | INFO | train rank=0 | step=57950/100000 progress=58.0% loss=0.043716 lr=4.103730e-06 elapsed=21:28:04 avg_step=1.33s recent_step=1.42s eta=15:34:39 +2026-08-26 01:23:49 | INFO | train rank=0 | step=57960/100000 progress=58.0% loss=0.077301 lr=4.102103e-06 elapsed=21:28:19 avg_step=1.33s recent_step=1.48s eta=15:34:27 +2026-08-26 01:24:01 | INFO | train rank=0 | step=57970/100000 progress=58.0% loss=0.042938 lr=4.100476e-06 elapsed=21:28:31 avg_step=1.33s recent_step=1.21s eta=15:34:13 +2026-08-26 01:24:15 | INFO | train rank=0 | step=57980/100000 progress=58.0% loss=0.057114 lr=4.098850e-06 elapsed=21:28:45 avg_step=1.33s recent_step=1.43s eta=15:34:00 +2026-08-26 01:24:32 | INFO | train rank=0 | step=57990/100000 progress=58.0% loss=0.058807 lr=4.097224e-06 elapsed=21:29:01 avg_step=1.33s recent_step=1.62s eta=15:33:49 +2026-08-26 01:24:43 | INFO | train rank=0 | step=58000/100000 progress=58.0% loss=0.031578 lr=4.095597e-06 elapsed=21:29:13 avg_step=1.33s recent_step=1.18s eta=15:33:34 +2026-08-26 01:24:56 | INFO | train rank=0 | step=58010/100000 progress=58.0% loss=0.065521 lr=4.093971e-06 elapsed=21:29:26 avg_step=1.33s recent_step=1.28s eta=15:33:21 +2026-08-26 01:25:08 | INFO | train rank=0 | step=58020/100000 progress=58.0% loss=0.072083 lr=4.092345e-06 elapsed=21:29:38 avg_step=1.33s recent_step=1.20s eta=15:33:06 +2026-08-26 01:25:24 | INFO | train rank=0 | step=58030/100000 progress=58.0% loss=0.066315 lr=4.090719e-06 elapsed=21:29:53 avg_step=1.33s recent_step=1.53s eta=15:32:54 +2026-08-26 01:25:39 | INFO | train rank=0 | step=58040/100000 progress=58.0% loss=0.052795 lr=4.089093e-06 elapsed=21:30:09 avg_step=1.33s recent_step=1.58s eta=15:32:43 +2026-08-26 01:25:51 | INFO | train rank=0 | step=58050/100000 progress=58.1% loss=0.065063 lr=4.087468e-06 elapsed=21:30:21 avg_step=1.33s recent_step=1.14s eta=15:32:28 +2026-08-26 01:26:03 | INFO | train rank=0 | step=58060/100000 progress=58.1% loss=0.051102 lr=4.085842e-06 elapsed=21:30:33 avg_step=1.33s recent_step=1.25s eta=15:32:14 +2026-08-26 01:26:18 | INFO | train rank=0 | step=58070/100000 progress=58.1% loss=0.059723 lr=4.084216e-06 elapsed=21:30:48 avg_step=1.33s recent_step=1.46s eta=15:32:02 +2026-08-26 01:26:29 | INFO | train rank=0 | step=58080/100000 progress=58.1% loss=0.057373 lr=4.082591e-06 elapsed=21:30:59 avg_step=1.33s recent_step=1.14s eta=15:31:47 +2026-08-26 01:26:44 | INFO | train rank=0 | step=58090/100000 progress=58.1% loss=0.046722 lr=4.080966e-06 elapsed=21:31:14 avg_step=1.33s recent_step=1.44s eta=15:31:34 +2026-08-26 01:26:56 | INFO | train rank=0 | step=58100/100000 progress=58.1% loss=0.097534 lr=4.079340e-06 elapsed=21:31:25 avg_step=1.33s recent_step=1.19s eta=15:31:20 +2026-08-26 01:27:07 | INFO | train rank=0 | step=58110/100000 progress=58.1% loss=0.048004 lr=4.077715e-06 elapsed=21:31:37 avg_step=1.33s recent_step=1.13s eta=15:31:05 +2026-08-26 01:27:21 | INFO | train rank=0 | step=58120/100000 progress=58.1% loss=0.077942 lr=4.076090e-06 elapsed=21:31:51 avg_step=1.33s recent_step=1.39s eta=15:30:52 +2026-08-26 01:27:33 | INFO | train rank=0 | step=58130/100000 progress=58.1% loss=0.033363 lr=4.074465e-06 elapsed=21:32:03 avg_step=1.33s recent_step=1.26s eta=15:30:38 +2026-08-26 01:27:48 | INFO | train rank=0 | step=58140/100000 progress=58.1% loss=0.065613 lr=4.072840e-06 elapsed=21:32:18 avg_step=1.33s recent_step=1.44s eta=15:30:26 +2026-08-26 01:28:02 | INFO | train rank=0 | step=58150/100000 progress=58.1% loss=0.058334 lr=4.071216e-06 elapsed=21:32:32 avg_step=1.33s recent_step=1.44s eta=15:30:13 +2026-08-26 01:28:16 | INFO | train rank=0 | step=58160/100000 progress=58.2% loss=0.053085 lr=4.069591e-06 elapsed=21:32:46 avg_step=1.33s recent_step=1.35s eta=15:30:00 +2026-08-26 01:28:27 | INFO | train rank=0 | step=58170/100000 progress=58.2% loss=0.076324 lr=4.067966e-06 elapsed=21:32:57 avg_step=1.33s recent_step=1.16s eta=15:29:46 +2026-08-26 01:28:39 | INFO | train rank=0 | step=58180/100000 progress=58.2% loss=0.066269 lr=4.066342e-06 elapsed=21:33:09 avg_step=1.33s recent_step=1.20s eta=15:29:31 +2026-08-26 01:28:51 | INFO | train rank=0 | step=58190/100000 progress=58.2% loss=0.035110 lr=4.064718e-06 elapsed=21:33:21 avg_step=1.33s recent_step=1.13s eta=15:29:17 +2026-08-26 01:29:06 | INFO | train rank=0 | step=58200/100000 progress=58.2% loss=0.103943 lr=4.063093e-06 elapsed=21:33:36 avg_step=1.33s recent_step=1.53s eta=15:29:05 +2026-08-26 01:29:22 | INFO | train rank=0 | step=58210/100000 progress=58.2% loss=0.058228 lr=4.061469e-06 elapsed=21:33:52 avg_step=1.33s recent_step=1.56s eta=15:28:53 +2026-08-26 01:29:33 | INFO | train rank=0 | step=58220/100000 progress=58.2% loss=0.040749 lr=4.059845e-06 elapsed=21:34:03 avg_step=1.33s recent_step=1.12s eta=15:28:38 +2026-08-26 01:29:49 | INFO | train rank=0 | step=58230/100000 progress=58.2% loss=0.049316 lr=4.058221e-06 elapsed=21:34:19 avg_step=1.33s recent_step=1.62s eta=15:28:27 +2026-08-26 01:30:01 | INFO | train rank=0 | step=58240/100000 progress=58.2% loss=0.049728 lr=4.056598e-06 elapsed=21:34:31 avg_step=1.33s recent_step=1.18s eta=15:28:12 +2026-08-26 01:30:16 | INFO | train rank=0 | step=58250/100000 progress=58.2% loss=0.020943 lr=4.054974e-06 elapsed=21:34:46 avg_step=1.33s recent_step=1.52s eta=15:28:00 +2026-08-26 01:30:28 | INFO | train rank=0 | step=58260/100000 progress=58.3% loss=0.086670 lr=4.053350e-06 elapsed=21:34:58 avg_step=1.33s recent_step=1.24s eta=15:27:46 +2026-08-26 01:30:45 | INFO | train rank=0 | step=58270/100000 progress=58.3% loss=0.055328 lr=4.051727e-06 elapsed=21:35:15 avg_step=1.33s recent_step=1.65s eta=15:27:35 +2026-08-26 01:30:59 | INFO | train rank=0 | step=58280/100000 progress=58.3% loss=0.043015 lr=4.050103e-06 elapsed=21:35:29 avg_step=1.33s recent_step=1.41s eta=15:27:22 +2026-08-26 01:31:13 | INFO | train rank=0 | step=58290/100000 progress=58.3% loss=0.046844 lr=4.048480e-06 elapsed=21:35:43 avg_step=1.33s recent_step=1.43s eta=15:27:10 +2026-08-26 01:31:26 | INFO | train rank=0 | step=58300/100000 progress=58.3% loss=0.065735 lr=4.046857e-06 elapsed=21:35:56 avg_step=1.33s recent_step=1.24s eta=15:26:56 +2026-08-26 01:31:38 | INFO | train rank=0 | step=58310/100000 progress=58.3% loss=0.042938 lr=4.045234e-06 elapsed=21:36:08 avg_step=1.33s recent_step=1.20s eta=15:26:42 +2026-08-26 01:31:50 | INFO | train rank=0 | step=58320/100000 progress=58.3% loss=0.041901 lr=4.043611e-06 elapsed=21:36:20 avg_step=1.33s recent_step=1.24s eta=15:26:27 +2026-08-26 01:32:02 | INFO | train rank=0 | step=58330/100000 progress=58.3% loss=0.046783 lr=4.041988e-06 elapsed=21:36:32 avg_step=1.33s recent_step=1.18s eta=15:26:13 +2026-08-26 01:32:15 | INFO | train rank=0 | step=58340/100000 progress=58.3% loss=0.043152 lr=4.040365e-06 elapsed=21:36:45 avg_step=1.33s recent_step=1.30s eta=15:25:59 +2026-08-26 01:32:29 | INFO | train rank=0 | step=58350/100000 progress=58.4% loss=0.114746 lr=4.038742e-06 elapsed=21:36:59 avg_step=1.33s recent_step=1.39s eta=15:25:46 +2026-08-26 01:32:42 | INFO | train rank=0 | step=58360/100000 progress=58.4% loss=0.057693 lr=4.037120e-06 elapsed=21:37:12 avg_step=1.33s recent_step=1.37s eta=15:25:33 +2026-08-26 01:32:54 | INFO | train rank=0 | step=58370/100000 progress=58.4% loss=0.046432 lr=4.035497e-06 elapsed=21:37:24 avg_step=1.33s recent_step=1.17s eta=15:25:19 +2026-08-26 01:33:08 | INFO | train rank=0 | step=58380/100000 progress=58.4% loss=0.039185 lr=4.033875e-06 elapsed=21:37:38 avg_step=1.33s recent_step=1.36s eta=15:25:06 +2026-08-26 01:33:29 | INFO | train rank=0 | step=58390/100000 progress=58.4% loss=0.039520 lr=4.032253e-06 elapsed=21:37:59 avg_step=1.33s recent_step=2.11s eta=15:24:58 +2026-08-26 01:33:40 | INFO | train rank=0 | step=58400/100000 progress=58.4% loss=0.034386 lr=4.030631e-06 elapsed=21:38:10 avg_step=1.33s recent_step=1.13s eta=15:24:43 +2026-08-26 01:33:55 | INFO | train rank=0 | step=58410/100000 progress=58.4% loss=0.045395 lr=4.029008e-06 elapsed=21:38:25 avg_step=1.33s recent_step=1.49s eta=15:24:31 +2026-08-26 01:34:10 | INFO | train rank=0 | step=58420/100000 progress=58.4% loss=0.032654 lr=4.027387e-06 elapsed=21:38:40 avg_step=1.33s recent_step=1.52s eta=15:24:19 +2026-08-26 01:34:27 | INFO | train rank=0 | step=58430/100000 progress=58.4% loss=0.068039 lr=4.025765e-06 elapsed=21:38:57 avg_step=1.33s recent_step=1.69s eta=15:24:08 +2026-08-26 01:34:40 | INFO | train rank=0 | step=58440/100000 progress=58.4% loss=0.057999 lr=4.024143e-06 elapsed=21:39:10 avg_step=1.33s recent_step=1.26s eta=15:23:54 +2026-08-26 01:34:51 | INFO | train rank=0 | step=58450/100000 progress=58.5% loss=0.047058 lr=4.022521e-06 elapsed=21:39:21 avg_step=1.33s recent_step=1.13s eta=15:23:40 +2026-08-26 01:35:03 | INFO | train rank=0 | step=58460/100000 progress=58.5% loss=0.038940 lr=4.020900e-06 elapsed=21:39:33 avg_step=1.33s recent_step=1.17s eta=15:23:25 +2026-08-26 01:35:14 | INFO | train rank=0 | step=58470/100000 progress=58.5% loss=0.066742 lr=4.019278e-06 elapsed=21:39:44 avg_step=1.33s recent_step=1.10s eta=15:23:10 +2026-08-26 01:35:26 | INFO | train rank=0 | step=58480/100000 progress=58.5% loss=0.052155 lr=4.017657e-06 elapsed=21:39:55 avg_step=1.33s recent_step=1.18s eta=15:22:56 +2026-08-26 01:35:37 | INFO | train rank=0 | step=58490/100000 progress=58.5% loss=0.048996 lr=4.016036e-06 elapsed=21:40:07 avg_step=1.33s recent_step=1.14s eta=15:22:41 +2026-08-26 01:35:49 | INFO | train rank=0 | step=58500/100000 progress=58.5% loss=0.062195 lr=4.014415e-06 elapsed=21:40:19 avg_step=1.33s recent_step=1.19s eta=15:22:27 +2026-08-26 01:36:07 | INFO | train rank=0 | step=58510/100000 progress=58.5% loss=0.050812 lr=4.012794e-06 elapsed=21:40:37 avg_step=1.33s recent_step=1.79s eta=15:22:16 +2026-08-26 01:36:21 | INFO | train rank=0 | step=58520/100000 progress=58.5% loss=0.049591 lr=4.011173e-06 elapsed=21:40:51 avg_step=1.33s recent_step=1.43s eta=15:22:04 +2026-08-26 01:36:34 | INFO | train rank=0 | step=58530/100000 progress=58.5% loss=0.049614 lr=4.009552e-06 elapsed=21:41:04 avg_step=1.33s recent_step=1.27s eta=15:21:50 +2026-08-26 01:36:46 | INFO | train rank=0 | step=58540/100000 progress=58.5% loss=0.046204 lr=4.007932e-06 elapsed=21:41:15 avg_step=1.33s recent_step=1.18s eta=15:21:36 +2026-08-26 01:37:00 | INFO | train rank=0 | step=58550/100000 progress=58.6% loss=0.071259 lr=4.006311e-06 elapsed=21:41:30 avg_step=1.33s recent_step=1.47s eta=15:21:23 +2026-08-26 01:37:12 | INFO | train rank=0 | step=58560/100000 progress=58.6% loss=0.041290 lr=4.004691e-06 elapsed=21:41:42 avg_step=1.33s recent_step=1.16s eta=15:21:09 +2026-08-26 01:37:25 | INFO | train rank=0 | step=58570/100000 progress=58.6% loss=0.080872 lr=4.003070e-06 elapsed=21:41:54 avg_step=1.33s recent_step=1.27s eta=15:20:55 +2026-08-26 01:37:39 | INFO | train rank=0 | step=58580/100000 progress=58.6% loss=0.055374 lr=4.001450e-06 elapsed=21:42:09 avg_step=1.33s recent_step=1.46s eta=15:20:42 +2026-08-26 01:37:54 | INFO | train rank=0 | step=58590/100000 progress=58.6% loss=0.044289 lr=3.999830e-06 elapsed=21:42:24 avg_step=1.33s recent_step=1.51s eta=15:20:30 +2026-08-26 01:38:10 | INFO | train rank=0 | step=58600/100000 progress=58.6% loss=0.051605 lr=3.998210e-06 elapsed=21:42:40 avg_step=1.33s recent_step=1.56s eta=15:20:18 +2026-08-26 01:38:26 | INFO | train rank=0 | step=58610/100000 progress=58.6% loss=0.033752 lr=3.996590e-06 elapsed=21:42:56 avg_step=1.33s recent_step=1.65s eta=15:20:07 +2026-08-26 01:38:41 | INFO | train rank=0 | step=58620/100000 progress=58.6% loss=0.121582 lr=3.994970e-06 elapsed=21:43:11 avg_step=1.33s recent_step=1.45s eta=15:19:55 +2026-08-26 01:38:54 | INFO | train rank=0 | step=58630/100000 progress=58.6% loss=0.079620 lr=3.993351e-06 elapsed=21:43:24 avg_step=1.33s recent_step=1.34s eta=15:19:42 +2026-08-26 01:39:07 | INFO | train rank=0 | step=58640/100000 progress=58.6% loss=0.045013 lr=3.991731e-06 elapsed=21:43:37 avg_step=1.33s recent_step=1.31s eta=15:19:28 +2026-08-26 01:39:21 | INFO | train rank=0 | step=58650/100000 progress=58.7% loss=0.055359 lr=3.990112e-06 elapsed=21:43:51 avg_step=1.33s recent_step=1.34s eta=15:19:15 +2026-08-26 01:39:37 | INFO | train rank=0 | step=58660/100000 progress=58.7% loss=0.039810 lr=3.988492e-06 elapsed=21:44:07 avg_step=1.33s recent_step=1.68s eta=15:19:04 +2026-08-26 01:39:49 | INFO | train rank=0 | step=58670/100000 progress=58.7% loss=0.060028 lr=3.986873e-06 elapsed=21:44:19 avg_step=1.33s recent_step=1.17s eta=15:18:49 +2026-08-26 01:40:01 | INFO | train rank=0 | step=58680/100000 progress=58.7% loss=0.055756 lr=3.985254e-06 elapsed=21:44:31 avg_step=1.33s recent_step=1.17s eta=15:18:35 +2026-08-26 01:40:13 | INFO | train rank=0 | step=58690/100000 progress=58.7% loss=0.051605 lr=3.983635e-06 elapsed=21:44:43 avg_step=1.33s recent_step=1.23s eta=15:18:21 +2026-08-26 01:40:28 | INFO | train rank=0 | step=58700/100000 progress=58.7% loss=0.062531 lr=3.982016e-06 elapsed=21:44:58 avg_step=1.33s recent_step=1.52s eta=15:18:09 +2026-08-26 01:40:45 | INFO | train rank=0 | step=58710/100000 progress=58.7% loss=0.074585 lr=3.980397e-06 elapsed=21:45:15 avg_step=1.33s recent_step=1.67s eta=15:17:58 +2026-08-26 01:40:57 | INFO | train rank=0 | step=58720/100000 progress=58.7% loss=0.066742 lr=3.978779e-06 elapsed=21:45:27 avg_step=1.33s recent_step=1.19s eta=15:17:43 +2026-08-26 01:41:11 | INFO | train rank=0 | step=58730/100000 progress=58.7% loss=0.047882 lr=3.977160e-06 elapsed=21:45:41 avg_step=1.33s recent_step=1.43s eta=15:17:31 +2026-08-26 01:41:27 | INFO | train rank=0 | step=58740/100000 progress=58.7% loss=0.030731 lr=3.975542e-06 elapsed=21:45:57 avg_step=1.33s recent_step=1.59s eta=15:17:19 +2026-08-26 01:41:38 | INFO | train rank=0 | step=58750/100000 progress=58.8% loss=0.038620 lr=3.973923e-06 elapsed=21:46:08 avg_step=1.33s recent_step=1.13s eta=15:17:04 +2026-08-26 01:41:56 | INFO | train rank=0 | step=58760/100000 progress=58.8% loss=0.045685 lr=3.972305e-06 elapsed=21:46:25 avg_step=1.33s recent_step=1.72s eta=15:16:54 +2026-08-26 01:42:08 | INFO | train rank=0 | step=58770/100000 progress=58.8% loss=0.036392 lr=3.970687e-06 elapsed=21:46:38 avg_step=1.33s recent_step=1.22s eta=15:16:40 +2026-08-26 01:42:19 | INFO | train rank=0 | step=58780/100000 progress=58.8% loss=0.055222 lr=3.969069e-06 elapsed=21:46:49 avg_step=1.33s recent_step=1.14s eta=15:16:25 +2026-08-26 01:42:34 | INFO | train rank=0 | step=58790/100000 progress=58.8% loss=0.037949 lr=3.967451e-06 elapsed=21:47:04 avg_step=1.33s recent_step=1.51s eta=15:16:13 +2026-08-26 01:42:47 | INFO | train rank=0 | step=58800/100000 progress=58.8% loss=0.059753 lr=3.965833e-06 elapsed=21:47:17 avg_step=1.33s recent_step=1.26s eta=15:15:59 +2026-08-26 01:43:01 | INFO | train rank=0 | step=58810/100000 progress=58.8% loss=0.059723 lr=3.964216e-06 elapsed=21:47:31 avg_step=1.33s recent_step=1.42s eta=15:15:46 +2026-08-26 01:43:16 | INFO | train rank=0 | step=58820/100000 progress=58.8% loss=0.047112 lr=3.962598e-06 elapsed=21:47:46 avg_step=1.33s recent_step=1.55s eta=15:15:34 +2026-08-26 01:43:31 | INFO | train rank=0 | step=58830/100000 progress=58.8% loss=0.038139 lr=3.960981e-06 elapsed=21:48:01 avg_step=1.33s recent_step=1.45s eta=15:15:22 +2026-08-26 01:43:45 | INFO | train rank=0 | step=58840/100000 progress=58.8% loss=0.027031 lr=3.959363e-06 elapsed=21:48:15 avg_step=1.33s recent_step=1.37s eta=15:15:09 +2026-08-26 01:43:57 | INFO | train rank=0 | step=58850/100000 progress=58.9% loss=0.046677 lr=3.957746e-06 elapsed=21:48:27 avg_step=1.33s recent_step=1.20s eta=15:14:54 +2026-08-26 01:44:10 | INFO | train rank=0 | step=58860/100000 progress=58.9% loss=0.051834 lr=3.956129e-06 elapsed=21:48:40 avg_step=1.33s recent_step=1.32s eta=15:14:41 +2026-08-26 01:44:22 | INFO | train rank=0 | step=58870/100000 progress=58.9% loss=0.025253 lr=3.954512e-06 elapsed=21:48:52 avg_step=1.33s recent_step=1.17s eta=15:14:27 +2026-08-26 01:44:34 | INFO | train rank=0 | step=58880/100000 progress=58.9% loss=0.059082 lr=3.952895e-06 elapsed=21:49:04 avg_step=1.33s recent_step=1.24s eta=15:14:13 +2026-08-26 01:44:45 | INFO | train rank=0 | step=58890/100000 progress=58.9% loss=0.039795 lr=3.951278e-06 elapsed=21:49:15 avg_step=1.33s recent_step=1.12s eta=15:13:58 +2026-08-26 01:45:02 | INFO | train rank=0 | step=58900/100000 progress=58.9% loss=0.063644 lr=3.949662e-06 elapsed=21:49:32 avg_step=1.33s recent_step=1.66s eta=15:13:47 +2026-08-26 01:45:14 | INFO | train rank=0 | step=58910/100000 progress=58.9% loss=0.038658 lr=3.948045e-06 elapsed=21:49:44 avg_step=1.33s recent_step=1.21s eta=15:13:32 +2026-08-26 01:45:27 | INFO | train rank=0 | step=58920/100000 progress=58.9% loss=0.035233 lr=3.946429e-06 elapsed=21:49:57 avg_step=1.33s recent_step=1.28s eta=15:13:19 +2026-08-26 01:45:40 | INFO | train rank=0 | step=58930/100000 progress=58.9% loss=0.047989 lr=3.944813e-06 elapsed=21:50:10 avg_step=1.33s recent_step=1.28s eta=15:13:05 +2026-08-26 01:45:55 | INFO | train rank=0 | step=58940/100000 progress=58.9% loss=0.043777 lr=3.943196e-06 elapsed=21:50:25 avg_step=1.33s recent_step=1.58s eta=15:12:53 +2026-08-26 01:46:10 | INFO | train rank=0 | step=58950/100000 progress=59.0% loss=0.053055 lr=3.941580e-06 elapsed=21:50:40 avg_step=1.33s recent_step=1.49s eta=15:12:41 +2026-08-26 01:46:27 | INFO | train rank=0 | step=58960/100000 progress=59.0% loss=0.054092 lr=3.939964e-06 elapsed=21:50:56 avg_step=1.33s recent_step=1.62s eta=15:12:30 +2026-08-26 01:46:41 | INFO | train rank=0 | step=58970/100000 progress=59.0% loss=0.092896 lr=3.938349e-06 elapsed=21:51:11 avg_step=1.33s recent_step=1.41s eta=15:12:17 +2026-08-26 01:46:55 | INFO | train rank=0 | step=58980/100000 progress=59.0% loss=0.055496 lr=3.936733e-06 elapsed=21:51:25 avg_step=1.33s recent_step=1.48s eta=15:12:05 +2026-08-26 01:47:12 | INFO | train rank=0 | step=58990/100000 progress=59.0% loss=0.041016 lr=3.935117e-06 elapsed=21:51:41 avg_step=1.33s recent_step=1.60s eta=15:11:53 +2026-08-26 01:47:23 | INFO | train rank=0 | step=59000/100000 progress=59.0% loss=0.048660 lr=3.933502e-06 elapsed=21:51:53 avg_step=1.33s recent_step=1.18s eta=15:11:39 +2026-08-26 01:47:38 | INFO | train rank=0 | step=59010/100000 progress=59.0% loss=0.038406 lr=3.931886e-06 elapsed=21:52:08 avg_step=1.33s recent_step=1.47s eta=15:11:26 +2026-08-26 01:47:49 | INFO | train rank=0 | step=59020/100000 progress=59.0% loss=0.047928 lr=3.930271e-06 elapsed=21:52:19 avg_step=1.33s recent_step=1.08s eta=15:11:11 +2026-08-26 01:48:05 | INFO | train rank=0 | step=59030/100000 progress=59.0% loss=0.061646 lr=3.928656e-06 elapsed=21:52:35 avg_step=1.33s recent_step=1.60s eta=15:11:00 +2026-08-26 01:48:18 | INFO | train rank=0 | step=59040/100000 progress=59.0% loss=0.049530 lr=3.927041e-06 elapsed=21:52:48 avg_step=1.33s recent_step=1.35s eta=15:10:47 +2026-08-26 01:48:31 | INFO | train rank=0 | step=59050/100000 progress=59.1% loss=0.043037 lr=3.925426e-06 elapsed=21:53:01 avg_step=1.33s recent_step=1.33s eta=15:10:33 +2026-08-26 01:48:43 | INFO | train rank=0 | step=59060/100000 progress=59.1% loss=0.071808 lr=3.923811e-06 elapsed=21:53:13 avg_step=1.33s recent_step=1.16s eta=15:10:19 +2026-08-26 01:48:59 | INFO | train rank=0 | step=59070/100000 progress=59.1% loss=0.056839 lr=3.922197e-06 elapsed=21:53:29 avg_step=1.33s recent_step=1.60s eta=15:10:07 +2026-08-26 01:49:14 | INFO | train rank=0 | step=59080/100000 progress=59.1% loss=0.043533 lr=3.920582e-06 elapsed=21:53:44 avg_step=1.33s recent_step=1.48s eta=15:09:55 +2026-08-26 01:49:26 | INFO | train rank=0 | step=59090/100000 progress=59.1% loss=0.062164 lr=3.918968e-06 elapsed=21:53:55 avg_step=1.33s recent_step=1.16s eta=15:09:40 +2026-08-26 01:49:37 | INFO | train rank=0 | step=59100/100000 progress=59.1% loss=0.066193 lr=3.917354e-06 elapsed=21:54:07 avg_step=1.33s recent_step=1.13s eta=15:09:25 +2026-08-26 01:49:48 | INFO | train rank=0 | step=59110/100000 progress=59.1% loss=0.048553 lr=3.915739e-06 elapsed=21:54:18 avg_step=1.33s recent_step=1.12s eta=15:09:11 +2026-08-26 01:50:03 | INFO | train rank=0 | step=59120/100000 progress=59.1% loss=0.034744 lr=3.914125e-06 elapsed=21:54:33 avg_step=1.33s recent_step=1.51s eta=15:08:59 +2026-08-26 01:50:17 | INFO | train rank=0 | step=59130/100000 progress=59.1% loss=0.060440 lr=3.912511e-06 elapsed=21:54:47 avg_step=1.33s recent_step=1.38s eta=15:08:46 +2026-08-26 01:50:31 | INFO | train rank=0 | step=59140/100000 progress=59.1% loss=0.050598 lr=3.910898e-06 elapsed=21:55:01 avg_step=1.33s recent_step=1.44s eta=15:08:33 +2026-08-26 01:50:43 | INFO | train rank=0 | step=59150/100000 progress=59.2% loss=0.037319 lr=3.909284e-06 elapsed=21:55:13 avg_step=1.33s recent_step=1.14s eta=15:08:18 +2026-08-26 01:50:58 | INFO | train rank=0 | step=59160/100000 progress=59.2% loss=0.060623 lr=3.907670e-06 elapsed=21:55:28 avg_step=1.33s recent_step=1.49s eta=15:08:06 +2026-08-26 01:51:09 | INFO | train rank=0 | step=59170/100000 progress=59.2% loss=0.078583 lr=3.906057e-06 elapsed=21:55:39 avg_step=1.33s recent_step=1.18s eta=15:07:52 +2026-08-26 01:51:21 | INFO | train rank=0 | step=59180/100000 progress=59.2% loss=0.040451 lr=3.904443e-06 elapsed=21:55:51 avg_step=1.33s recent_step=1.15s eta=15:07:37 +2026-08-26 01:51:38 | INFO | train rank=0 | step=59190/100000 progress=59.2% loss=0.081482 lr=3.902830e-06 elapsed=21:56:07 avg_step=1.33s recent_step=1.66s eta=15:07:26 +2026-08-26 01:51:50 | INFO | train rank=0 | step=59200/100000 progress=59.2% loss=0.055328 lr=3.901217e-06 elapsed=21:56:20 avg_step=1.33s recent_step=1.24s eta=15:07:12 +2026-08-26 01:52:02 | INFO | train rank=0 | step=59210/100000 progress=59.2% loss=0.030029 lr=3.899604e-06 elapsed=21:56:32 avg_step=1.33s recent_step=1.18s eta=15:06:58 +2026-08-26 01:52:19 | INFO | train rank=0 | step=59220/100000 progress=59.2% loss=0.062225 lr=3.897991e-06 elapsed=21:56:49 avg_step=1.33s recent_step=1.71s eta=15:06:47 +2026-08-26 01:52:35 | INFO | train rank=0 | step=59230/100000 progress=59.2% loss=0.053040 lr=3.896378e-06 elapsed=21:57:05 avg_step=1.33s recent_step=1.60s eta=15:06:35 +2026-08-26 01:52:47 | INFO | train rank=0 | step=59240/100000 progress=59.2% loss=0.067123 lr=3.894766e-06 elapsed=21:57:17 avg_step=1.33s recent_step=1.19s eta=15:06:21 +2026-08-26 01:52:58 | INFO | train rank=0 | step=59250/100000 progress=59.2% loss=0.026855 lr=3.893153e-06 elapsed=21:57:28 avg_step=1.33s recent_step=1.16s eta=15:06:06 +2026-08-26 01:53:11 | INFO | train rank=0 | step=59260/100000 progress=59.3% loss=0.055725 lr=3.891541e-06 elapsed=21:57:41 avg_step=1.33s recent_step=1.23s eta=15:05:52 +2026-08-26 01:53:27 | INFO | train rank=0 | step=59270/100000 progress=59.3% loss=0.043953 lr=3.889929e-06 elapsed=21:57:57 avg_step=1.33s recent_step=1.60s eta=15:05:41 +2026-08-26 01:53:42 | INFO | train rank=0 | step=59280/100000 progress=59.3% loss=0.050446 lr=3.888317e-06 elapsed=21:58:12 avg_step=1.33s recent_step=1.55s eta=15:05:29 +2026-08-26 01:53:56 | INFO | train rank=0 | step=59290/100000 progress=59.3% loss=0.045105 lr=3.886704e-06 elapsed=21:58:26 avg_step=1.33s recent_step=1.40s eta=15:05:16 +2026-08-26 01:54:10 | INFO | train rank=0 | step=59300/100000 progress=59.3% loss=0.028885 lr=3.885093e-06 elapsed=21:58:40 avg_step=1.33s recent_step=1.39s eta=15:05:03 +2026-08-26 01:54:24 | INFO | train rank=0 | step=59310/100000 progress=59.3% loss=0.050537 lr=3.883481e-06 elapsed=21:58:54 avg_step=1.33s recent_step=1.39s eta=15:04:50 +2026-08-26 01:54:39 | INFO | train rank=0 | step=59320/100000 progress=59.3% loss=0.061554 lr=3.881869e-06 elapsed=21:59:09 avg_step=1.33s recent_step=1.51s eta=15:04:38 +2026-08-26 01:54:51 | INFO | train rank=0 | step=59330/100000 progress=59.3% loss=0.147522 lr=3.880258e-06 elapsed=21:59:21 avg_step=1.33s recent_step=1.17s eta=15:04:24 +2026-08-26 01:55:06 | INFO | train rank=0 | step=59340/100000 progress=59.3% loss=0.053101 lr=3.878646e-06 elapsed=21:59:36 avg_step=1.33s recent_step=1.51s eta=15:04:11 +2026-08-26 01:55:18 | INFO | train rank=0 | step=59350/100000 progress=59.4% loss=0.034634 lr=3.877035e-06 elapsed=21:59:48 avg_step=1.33s recent_step=1.19s eta=15:03:57 +2026-08-26 01:55:30 | INFO | train rank=0 | step=59360/100000 progress=59.4% loss=0.096680 lr=3.875424e-06 elapsed=22:00:00 avg_step=1.33s recent_step=1.19s eta=15:03:43 +2026-08-26 01:55:42 | INFO | train rank=0 | step=59370/100000 progress=59.4% loss=0.027618 lr=3.873813e-06 elapsed=22:00:12 avg_step=1.33s recent_step=1.24s eta=15:03:29 +2026-08-26 01:55:59 | INFO | train rank=0 | step=59380/100000 progress=59.4% loss=0.054855 lr=3.872202e-06 elapsed=22:00:29 avg_step=1.33s recent_step=1.74s eta=15:03:18 +2026-08-26 01:56:15 | INFO | train rank=0 | step=59390/100000 progress=59.4% loss=0.037582 lr=3.870591e-06 elapsed=22:00:45 avg_step=1.33s recent_step=1.53s eta=15:03:06 +2026-08-26 01:56:26 | INFO | train rank=0 | step=59400/100000 progress=59.4% loss=0.033699 lr=3.868980e-06 elapsed=22:00:56 avg_step=1.33s recent_step=1.17s eta=15:02:52 +2026-08-26 01:56:38 | INFO | train rank=0 | step=59410/100000 progress=59.4% loss=0.061981 lr=3.867370e-06 elapsed=22:01:08 avg_step=1.33s recent_step=1.15s eta=15:02:37 +2026-08-26 01:56:55 | INFO | train rank=0 | step=59420/100000 progress=59.4% loss=0.038055 lr=3.865759e-06 elapsed=22:01:25 avg_step=1.33s recent_step=1.68s eta=15:02:26 +2026-08-26 01:57:07 | INFO | train rank=0 | step=59430/100000 progress=59.4% loss=0.044800 lr=3.864149e-06 elapsed=22:01:37 avg_step=1.33s recent_step=1.19s eta=15:02:12 +2026-08-26 01:57:20 | INFO | train rank=0 | step=59440/100000 progress=59.4% loss=0.036400 lr=3.862539e-06 elapsed=22:01:49 avg_step=1.33s recent_step=1.29s eta=15:01:58 +2026-08-26 01:57:32 | INFO | train rank=0 | step=59450/100000 progress=59.5% loss=0.030861 lr=3.860929e-06 elapsed=22:02:02 avg_step=1.33s recent_step=1.21s eta=15:01:44 +2026-08-26 01:57:46 | INFO | train rank=0 | step=59460/100000 progress=59.5% loss=0.036606 lr=3.859319e-06 elapsed=22:02:16 avg_step=1.33s recent_step=1.44s eta=15:01:31 +2026-08-26 01:57:58 | INFO | train rank=0 | step=59470/100000 progress=59.5% loss=0.062256 lr=3.857709e-06 elapsed=22:02:28 avg_step=1.33s recent_step=1.22s eta=15:01:17 +2026-08-26 01:58:13 | INFO | train rank=0 | step=59480/100000 progress=59.5% loss=0.081299 lr=3.856099e-06 elapsed=22:02:43 avg_step=1.33s recent_step=1.50s eta=15:01:05 +2026-08-26 01:58:26 | INFO | train rank=0 | step=59490/100000 progress=59.5% loss=0.036964 lr=3.854490e-06 elapsed=22:02:56 avg_step=1.33s recent_step=1.26s eta=15:00:51 +2026-08-26 01:58:40 | INFO | train rank=0 | step=59500/100000 progress=59.5% loss=0.059479 lr=3.852880e-06 elapsed=22:03:09 avg_step=1.33s recent_step=1.36s eta=15:00:38 +2026-08-26 01:58:53 | INFO | train rank=0 | step=59510/100000 progress=59.5% loss=0.059174 lr=3.851271e-06 elapsed=22:03:22 avg_step=1.33s recent_step=1.29s eta=15:00:24 +2026-08-26 01:59:04 | INFO | train rank=0 | step=59520/100000 progress=59.5% loss=0.040932 lr=3.849662e-06 elapsed=22:03:34 avg_step=1.33s recent_step=1.12s eta=15:00:10 +2026-08-26 01:59:16 | INFO | train rank=0 | step=59530/100000 progress=59.5% loss=0.075714 lr=3.848053e-06 elapsed=22:03:46 avg_step=1.33s recent_step=1.23s eta=14:59:56 +2026-08-26 01:59:30 | INFO | train rank=0 | step=59540/100000 progress=59.5% loss=0.076691 lr=3.846444e-06 elapsed=22:04:00 avg_step=1.33s recent_step=1.43s eta=14:59:43 +2026-08-26 01:59:47 | INFO | train rank=0 | step=59550/100000 progress=59.6% loss=0.056564 lr=3.844835e-06 elapsed=22:04:17 avg_step=1.33s recent_step=1.66s eta=14:59:32 +2026-08-26 01:59:59 | INFO | train rank=0 | step=59560/100000 progress=59.6% loss=0.033066 lr=3.843226e-06 elapsed=22:04:29 avg_step=1.33s recent_step=1.18s eta=14:59:17 +2026-08-26 02:00:13 | INFO | train rank=0 | step=59570/100000 progress=59.6% loss=0.032837 lr=3.841618e-06 elapsed=22:04:43 avg_step=1.33s recent_step=1.45s eta=14:59:05 +2026-08-26 02:00:27 | INFO | train rank=0 | step=59580/100000 progress=59.6% loss=0.051941 lr=3.840009e-06 elapsed=22:04:57 avg_step=1.33s recent_step=1.38s eta=14:58:52 +2026-08-26 02:00:39 | INFO | train rank=0 | step=59590/100000 progress=59.6% loss=0.048172 lr=3.838401e-06 elapsed=22:05:09 avg_step=1.33s recent_step=1.23s eta=14:58:38 +2026-08-26 02:00:54 | INFO | train rank=0 | step=59600/100000 progress=59.6% loss=0.072693 lr=3.836793e-06 elapsed=22:05:24 avg_step=1.33s recent_step=1.42s eta=14:58:25 +2026-08-26 02:01:06 | INFO | train rank=0 | step=59610/100000 progress=59.6% loss=0.039146 lr=3.835185e-06 elapsed=22:05:36 avg_step=1.33s recent_step=1.26s eta=14:58:11 +2026-08-26 02:01:23 | INFO | train rank=0 | step=59620/100000 progress=59.6% loss=0.044342 lr=3.833577e-06 elapsed=22:05:53 avg_step=1.33s recent_step=1.69s eta=14:58:00 +2026-08-26 02:01:35 | INFO | train rank=0 | step=59630/100000 progress=59.6% loss=0.149963 lr=3.831969e-06 elapsed=22:06:05 avg_step=1.33s recent_step=1.19s eta=14:57:46 +2026-08-26 02:01:46 | INFO | train rank=0 | step=59640/100000 progress=59.6% loss=0.046234 lr=3.830362e-06 elapsed=22:06:16 avg_step=1.33s recent_step=1.16s eta=14:57:31 +2026-08-26 02:01:58 | INFO | train rank=0 | step=59650/100000 progress=59.7% loss=0.056305 lr=3.828754e-06 elapsed=22:06:28 avg_step=1.33s recent_step=1.17s eta=14:57:17 +2026-08-26 02:02:15 | INFO | train rank=0 | step=59660/100000 progress=59.7% loss=0.059753 lr=3.827147e-06 elapsed=22:06:45 avg_step=1.33s recent_step=1.66s eta=14:57:06 +2026-08-26 02:02:30 | INFO | train rank=0 | step=59670/100000 progress=59.7% loss=0.051483 lr=3.825539e-06 elapsed=22:07:00 avg_step=1.33s recent_step=1.57s eta=14:56:54 +2026-08-26 02:02:42 | INFO | train rank=0 | step=59680/100000 progress=59.7% loss=0.060623 lr=3.823932e-06 elapsed=22:07:12 avg_step=1.33s recent_step=1.16s eta=14:56:39 +2026-08-26 02:02:54 | INFO | train rank=0 | step=59690/100000 progress=59.7% loss=0.049484 lr=3.822325e-06 elapsed=22:07:24 avg_step=1.33s recent_step=1.18s eta=14:56:25 +2026-08-26 02:03:09 | INFO | train rank=0 | step=59700/100000 progress=59.7% loss=0.037178 lr=3.820718e-06 elapsed=22:07:39 avg_step=1.33s recent_step=1.51s eta=14:56:13 +2026-08-26 02:03:21 | INFO | train rank=0 | step=59710/100000 progress=59.7% loss=0.038696 lr=3.819112e-06 elapsed=22:07:51 avg_step=1.33s recent_step=1.19s eta=14:55:59 +2026-08-26 02:03:34 | INFO | train rank=0 | step=59720/100000 progress=59.7% loss=0.038757 lr=3.817505e-06 elapsed=22:08:04 avg_step=1.33s recent_step=1.36s eta=14:55:45 +2026-08-26 02:03:46 | INFO | train rank=0 | step=59730/100000 progress=59.7% loss=0.042709 lr=3.815899e-06 elapsed=22:08:16 avg_step=1.33s recent_step=1.20s eta=14:55:31 +2026-08-26 02:04:03 | INFO | train rank=0 | step=59740/100000 progress=59.7% loss=0.050827 lr=3.814292e-06 elapsed=22:08:33 avg_step=1.33s recent_step=1.65s eta=14:55:20 +2026-08-26 02:04:15 | INFO | train rank=0 | step=59750/100000 progress=59.8% loss=0.052826 lr=3.812686e-06 elapsed=22:08:45 avg_step=1.33s recent_step=1.23s eta=14:55:06 +2026-08-26 02:04:33 | INFO | train rank=0 | step=59760/100000 progress=59.8% loss=0.091431 lr=3.811080e-06 elapsed=22:09:03 avg_step=1.33s recent_step=1.82s eta=14:54:56 +2026-08-26 02:04:44 | INFO | train rank=0 | step=59770/100000 progress=59.8% loss=0.034142 lr=3.809474e-06 elapsed=22:09:14 avg_step=1.33s recent_step=1.12s eta=14:54:41 +2026-08-26 02:05:00 | INFO | train rank=0 | step=59780/100000 progress=59.8% loss=0.029221 lr=3.807868e-06 elapsed=22:09:30 avg_step=1.33s recent_step=1.53s eta=14:54:29 +2026-08-26 02:05:12 | INFO | train rank=0 | step=59790/100000 progress=59.8% loss=0.064484 lr=3.806262e-06 elapsed=22:09:41 avg_step=1.33s recent_step=1.18s eta=14:54:14 +2026-08-26 02:05:23 | INFO | train rank=0 | step=59800/100000 progress=59.8% loss=0.112427 lr=3.804657e-06 elapsed=22:09:53 avg_step=1.33s recent_step=1.16s eta=14:54:00 +2026-08-26 02:05:35 | INFO | train rank=0 | step=59810/100000 progress=59.8% loss=0.043625 lr=3.803051e-06 elapsed=22:10:05 avg_step=1.33s recent_step=1.19s eta=14:53:46 +2026-08-26 02:05:47 | INFO | train rank=0 | step=59820/100000 progress=59.8% loss=0.031265 lr=3.801446e-06 elapsed=22:10:17 avg_step=1.33s recent_step=1.18s eta=14:53:31 +2026-08-26 02:06:04 | INFO | train rank=0 | step=59830/100000 progress=59.8% loss=0.038689 lr=3.799841e-06 elapsed=22:10:33 avg_step=1.33s recent_step=1.67s eta=14:53:20 +2026-08-26 02:06:15 | INFO | train rank=0 | step=59840/100000 progress=59.8% loss=0.085876 lr=3.798236e-06 elapsed=22:10:45 avg_step=1.33s recent_step=1.17s eta=14:53:06 +2026-08-26 02:06:27 | INFO | train rank=0 | step=59850/100000 progress=59.9% loss=0.055206 lr=3.796631e-06 elapsed=22:10:56 avg_step=1.33s recent_step=1.14s eta=14:52:51 +2026-08-26 02:06:46 | INFO | train rank=0 | step=59860/100000 progress=59.9% loss=0.052170 lr=3.795026e-06 elapsed=22:11:16 avg_step=1.33s recent_step=1.93s eta=14:52:42 +2026-08-26 02:06:58 | INFO | train rank=0 | step=59870/100000 progress=59.9% loss=0.056915 lr=3.793421e-06 elapsed=22:11:27 avg_step=1.33s recent_step=1.16s eta=14:52:27 +2026-08-26 02:07:12 | INFO | train rank=0 | step=59880/100000 progress=59.9% loss=0.045059 lr=3.791817e-06 elapsed=22:11:42 avg_step=1.33s recent_step=1.43s eta=14:52:15 +2026-08-26 02:07:29 | INFO | train rank=0 | step=59890/100000 progress=59.9% loss=0.039642 lr=3.790212e-06 elapsed=22:11:59 avg_step=1.33s recent_step=1.74s eta=14:52:04 +2026-08-26 02:07:41 | INFO | train rank=0 | step=59900/100000 progress=59.9% loss=0.021915 lr=3.788608e-06 elapsed=22:12:11 avg_step=1.33s recent_step=1.16s eta=14:51:49 +2026-08-26 02:07:52 | INFO | train rank=0 | step=59910/100000 progress=59.9% loss=0.041618 lr=3.787004e-06 elapsed=22:12:22 avg_step=1.33s recent_step=1.15s eta=14:51:35 +2026-08-26 02:08:04 | INFO | train rank=0 | step=59920/100000 progress=59.9% loss=0.071777 lr=3.785400e-06 elapsed=22:12:34 avg_step=1.33s recent_step=1.18s eta=14:51:21 +2026-08-26 02:08:17 | INFO | train rank=0 | step=59930/100000 progress=59.9% loss=0.038025 lr=3.783796e-06 elapsed=22:12:47 avg_step=1.33s recent_step=1.24s eta=14:51:07 +2026-08-26 02:08:33 | INFO | train rank=0 | step=59940/100000 progress=59.9% loss=0.058655 lr=3.782192e-06 elapsed=22:13:03 avg_step=1.33s recent_step=1.68s eta=14:50:55 +2026-08-26 02:08:45 | INFO | train rank=0 | step=59950/100000 progress=60.0% loss=0.060120 lr=3.780589e-06 elapsed=22:13:15 avg_step=1.33s recent_step=1.16s eta=14:50:41 +2026-08-26 02:08:59 | INFO | train rank=0 | step=59960/100000 progress=60.0% loss=0.069824 lr=3.778985e-06 elapsed=22:13:28 avg_step=1.33s recent_step=1.35s eta=14:50:28 +2026-08-26 02:09:10 | INFO | train rank=0 | step=59970/100000 progress=60.0% loss=0.052139 lr=3.777382e-06 elapsed=22:13:40 avg_step=1.33s recent_step=1.14s eta=14:50:13 +2026-08-26 02:09:22 | INFO | train rank=0 | step=59980/100000 progress=60.0% loss=0.059006 lr=3.775779e-06 elapsed=22:13:52 avg_step=1.33s recent_step=1.21s eta=14:49:59 +2026-08-26 02:09:34 | INFO | train rank=0 | step=59990/100000 progress=60.0% loss=0.049942 lr=3.774176e-06 elapsed=22:14:03 avg_step=1.33s recent_step=1.16s eta=14:49:44 +2026-08-26 02:09:48 | INFO | train rank=0 | step=60000/100000 progress=60.0% loss=0.074402 lr=3.772573e-06 elapsed=22:14:18 avg_step=1.33s recent_step=1.46s eta=14:49:32 +2026-08-26 02:10:04 | INFO | train rank=0 | step=60010/100000 progress=60.0% loss=0.050446 lr=3.770970e-06 elapsed=22:14:34 avg_step=1.33s recent_step=1.55s eta=14:49:20 +2026-08-26 02:10:20 | INFO | train rank=0 | step=60020/100000 progress=60.0% loss=0.041382 lr=3.769367e-06 elapsed=22:14:50 avg_step=1.33s recent_step=1.67s eta=14:49:09 +2026-08-26 02:10:32 | INFO | train rank=0 | step=60030/100000 progress=60.0% loss=0.038933 lr=3.767765e-06 elapsed=22:15:02 avg_step=1.33s recent_step=1.18s eta=14:48:54 +2026-08-26 02:10:47 | INFO | train rank=0 | step=60040/100000 progress=60.0% loss=0.090881 lr=3.766162e-06 elapsed=22:15:17 avg_step=1.33s recent_step=1.46s eta=14:48:42 +2026-08-26 02:10:59 | INFO | train rank=0 | step=60050/100000 progress=60.1% loss=0.035065 lr=3.764560e-06 elapsed=22:15:29 avg_step=1.33s recent_step=1.23s eta=14:48:28 +2026-08-26 02:11:11 | INFO | train rank=0 | step=60060/100000 progress=60.1% loss=0.072968 lr=3.762958e-06 elapsed=22:15:40 avg_step=1.33s recent_step=1.15s eta=14:48:13 +2026-08-26 02:11:22 | INFO | train rank=0 | step=60070/100000 progress=60.1% loss=0.041000 lr=3.761356e-06 elapsed=22:15:52 avg_step=1.33s recent_step=1.16s eta=14:47:59 +2026-08-26 02:11:34 | INFO | train rank=0 | step=60080/100000 progress=60.1% loss=0.049301 lr=3.759754e-06 elapsed=22:16:04 avg_step=1.33s recent_step=1.18s eta=14:47:45 +2026-08-26 02:11:46 | INFO | train rank=0 | step=60090/100000 progress=60.1% loss=0.062103 lr=3.758152e-06 elapsed=22:16:15 avg_step=1.33s recent_step=1.15s eta=14:47:30 +2026-08-26 02:11:59 | INFO | train rank=0 | step=60100/100000 progress=60.1% loss=0.048782 lr=3.756551e-06 elapsed=22:16:28 avg_step=1.33s recent_step=1.30s eta=14:47:16 +2026-08-26 02:12:14 | INFO | train rank=0 | step=60110/100000 progress=60.1% loss=0.046494 lr=3.754949e-06 elapsed=22:16:43 avg_step=1.33s recent_step=1.50s eta=14:47:04 +2026-08-26 02:12:30 | INFO | train rank=0 | step=60120/100000 progress=60.1% loss=0.043777 lr=3.753348e-06 elapsed=22:17:00 avg_step=1.33s recent_step=1.66s eta=14:46:53 +2026-08-26 02:12:42 | INFO | train rank=0 | step=60130/100000 progress=60.1% loss=0.040573 lr=3.751747e-06 elapsed=22:17:12 avg_step=1.33s recent_step=1.17s eta=14:46:39 +2026-08-26 02:12:59 | INFO | train rank=0 | step=60140/100000 progress=60.1% loss=0.031654 lr=3.750146e-06 elapsed=22:17:29 avg_step=1.33s recent_step=1.68s eta=14:46:28 +2026-08-26 02:13:11 | INFO | train rank=0 | step=60150/100000 progress=60.2% loss=0.041458 lr=3.748545e-06 elapsed=22:17:40 avg_step=1.33s recent_step=1.18s eta=14:46:13 +2026-08-26 02:13:24 | INFO | train rank=0 | step=60160/100000 progress=60.2% loss=0.056000 lr=3.746944e-06 elapsed=22:17:54 avg_step=1.33s recent_step=1.35s eta=14:46:00 +2026-08-26 02:13:38 | INFO | train rank=0 | step=60170/100000 progress=60.2% loss=0.040787 lr=3.745343e-06 elapsed=22:18:08 avg_step=1.33s recent_step=1.41s eta=14:45:47 +2026-08-26 02:13:51 | INFO | train rank=0 | step=60180/100000 progress=60.2% loss=0.044113 lr=3.743743e-06 elapsed=22:18:21 avg_step=1.33s recent_step=1.29s eta=14:45:33 +2026-08-26 02:14:03 | INFO | train rank=0 | step=60190/100000 progress=60.2% loss=0.031235 lr=3.742142e-06 elapsed=22:18:33 avg_step=1.33s recent_step=1.17s eta=14:45:19 +2026-08-26 02:14:14 | INFO | train rank=0 | step=60200/100000 progress=60.2% loss=0.040009 lr=3.740542e-06 elapsed=22:18:44 avg_step=1.33s recent_step=1.15s eta=14:45:04 +2026-08-26 02:14:29 | INFO | train rank=0 | step=60210/100000 progress=60.2% loss=0.028221 lr=3.738942e-06 elapsed=22:18:59 avg_step=1.33s recent_step=1.47s eta=14:44:52 +2026-08-26 02:14:46 | INFO | train rank=0 | step=60220/100000 progress=60.2% loss=0.048615 lr=3.737342e-06 elapsed=22:19:16 avg_step=1.33s recent_step=1.72s eta=14:44:41 +2026-08-26 02:14:57 | INFO | train rank=0 | step=60230/100000 progress=60.2% loss=0.065186 lr=3.735742e-06 elapsed=22:19:27 avg_step=1.33s recent_step=1.12s eta=14:44:26 +2026-08-26 02:15:11 | INFO | train rank=0 | step=60240/100000 progress=60.2% loss=0.040894 lr=3.734143e-06 elapsed=22:19:41 avg_step=1.33s recent_step=1.42s eta=14:44:14 +2026-08-26 02:15:27 | INFO | train rank=0 | step=60250/100000 progress=60.2% loss=0.051834 lr=3.732543e-06 elapsed=22:19:57 avg_step=1.33s recent_step=1.57s eta=14:44:02 +2026-08-26 02:15:40 | INFO | train rank=0 | step=60260/100000 progress=60.3% loss=0.039413 lr=3.730944e-06 elapsed=22:20:10 avg_step=1.33s recent_step=1.27s eta=14:43:48 +2026-08-26 02:15:56 | INFO | train rank=0 | step=60270/100000 progress=60.3% loss=0.070007 lr=3.729344e-06 elapsed=22:20:26 avg_step=1.33s recent_step=1.61s eta=14:43:37 +2026-08-26 02:16:07 | INFO | train rank=0 | step=60280/100000 progress=60.3% loss=0.048645 lr=3.727745e-06 elapsed=22:20:37 avg_step=1.33s recent_step=1.12s eta=14:43:22 +2026-08-26 02:16:19 | INFO | train rank=0 | step=60290/100000 progress=60.3% loss=0.040573 lr=3.726146e-06 elapsed=22:20:49 avg_step=1.33s recent_step=1.18s eta=14:43:07 +2026-08-26 02:16:32 | INFO | train rank=0 | step=60300/100000 progress=60.3% loss=0.074493 lr=3.724548e-06 elapsed=22:21:02 avg_step=1.33s recent_step=1.29s eta=14:42:54 +2026-08-26 02:16:47 | INFO | train rank=0 | step=60310/100000 progress=60.3% loss=0.051590 lr=3.722949e-06 elapsed=22:21:17 avg_step=1.33s recent_step=1.53s eta=14:42:42 +2026-08-26 02:16:59 | INFO | train rank=0 | step=60320/100000 progress=60.3% loss=0.020523 lr=3.721350e-06 elapsed=22:21:29 avg_step=1.33s recent_step=1.18s eta=14:42:27 +2026-08-26 02:17:13 | INFO | train rank=0 | step=60330/100000 progress=60.3% loss=0.052166 lr=3.719752e-06 elapsed=22:21:43 avg_step=1.33s recent_step=1.43s eta=14:42:15 +2026-08-26 02:17:27 | INFO | train rank=0 | step=60340/100000 progress=60.3% loss=0.037117 lr=3.718154e-06 elapsed=22:21:57 avg_step=1.33s recent_step=1.39s eta=14:42:02 +2026-08-26 02:17:40 | INFO | train rank=0 | step=60350/100000 progress=60.4% loss=0.052536 lr=3.716555e-06 elapsed=22:22:10 avg_step=1.33s recent_step=1.27s eta=14:41:48 +2026-08-26 02:17:53 | INFO | train rank=0 | step=60360/100000 progress=60.4% loss=0.046005 lr=3.714957e-06 elapsed=22:22:23 avg_step=1.33s recent_step=1.35s eta=14:41:35 +2026-08-26 02:18:09 | INFO | train rank=0 | step=60370/100000 progress=60.4% loss=0.062820 lr=3.713360e-06 elapsed=22:22:38 avg_step=1.33s recent_step=1.52s eta=14:41:23 +2026-08-26 02:18:22 | INFO | train rank=0 | step=60380/100000 progress=60.4% loss=0.034119 lr=3.711762e-06 elapsed=22:22:52 avg_step=1.33s recent_step=1.38s eta=14:41:10 +2026-08-26 02:18:34 | INFO | train rank=0 | step=60390/100000 progress=60.4% loss=0.047607 lr=3.710164e-06 elapsed=22:23:04 avg_step=1.33s recent_step=1.15s eta=14:40:55 +2026-08-26 02:18:45 | INFO | train rank=0 | step=60400/100000 progress=60.4% loss=0.038902 lr=3.708567e-06 elapsed=22:23:15 avg_step=1.33s recent_step=1.16s eta=14:40:40 +2026-08-26 02:18:57 | INFO | train rank=0 | step=60410/100000 progress=60.4% loss=0.051086 lr=3.706970e-06 elapsed=22:23:27 avg_step=1.33s recent_step=1.12s eta=14:40:26 +2026-08-26 02:19:11 | INFO | train rank=0 | step=60420/100000 progress=60.4% loss=0.083679 lr=3.705372e-06 elapsed=22:23:41 avg_step=1.33s recent_step=1.47s eta=14:40:13 +2026-08-26 02:19:23 | INFO | train rank=0 | step=60430/100000 progress=60.4% loss=0.039665 lr=3.703775e-06 elapsed=22:23:53 avg_step=1.33s recent_step=1.14s eta=14:39:59 +2026-08-26 02:19:38 | INFO | train rank=0 | step=60440/100000 progress=60.4% loss=0.057571 lr=3.702179e-06 elapsed=22:24:08 avg_step=1.33s recent_step=1.57s eta=14:39:47 +2026-08-26 02:19:50 | INFO | train rank=0 | step=60450/100000 progress=60.5% loss=0.068176 lr=3.700582e-06 elapsed=22:24:20 avg_step=1.33s recent_step=1.19s eta=14:39:33 +2026-08-26 02:20:05 | INFO | train rank=0 | step=60460/100000 progress=60.5% loss=0.044205 lr=3.698985e-06 elapsed=22:24:35 avg_step=1.33s recent_step=1.47s eta=14:39:20 +2026-08-26 02:20:17 | INFO | train rank=0 | step=60470/100000 progress=60.5% loss=0.066132 lr=3.697389e-06 elapsed=22:24:46 avg_step=1.33s recent_step=1.14s eta=14:39:06 +2026-08-26 02:20:28 | INFO | train rank=0 | step=60480/100000 progress=60.5% loss=0.020218 lr=3.695792e-06 elapsed=22:24:58 avg_step=1.33s recent_step=1.18s eta=14:38:51 +2026-08-26 02:20:40 | INFO | train rank=0 | step=60490/100000 progress=60.5% loss=0.037781 lr=3.694196e-06 elapsed=22:25:10 avg_step=1.33s recent_step=1.16s eta=14:38:37 +2026-08-26 02:20:58 | INFO | train rank=0 | step=60500/100000 progress=60.5% loss=0.041321 lr=3.692600e-06 elapsed=22:25:28 avg_step=1.33s recent_step=1.85s eta=14:38:27 +2026-08-26 02:21:11 | INFO | train rank=0 | step=60510/100000 progress=60.5% loss=0.037674 lr=3.691004e-06 elapsed=22:25:40 avg_step=1.33s recent_step=1.21s eta=14:38:13 +2026-08-26 02:21:26 | INFO | train rank=0 | step=60520/100000 progress=60.5% loss=0.073364 lr=3.689409e-06 elapsed=22:25:55 avg_step=1.33s recent_step=1.50s eta=14:38:00 +2026-08-26 02:21:38 | INFO | train rank=0 | step=60530/100000 progress=60.5% loss=0.027245 lr=3.687813e-06 elapsed=22:26:07 avg_step=1.33s recent_step=1.19s eta=14:37:46 +2026-08-26 02:21:49 | INFO | train rank=0 | step=60540/100000 progress=60.5% loss=0.035805 lr=3.686218e-06 elapsed=22:26:19 avg_step=1.33s recent_step=1.19s eta=14:37:32 +2026-08-26 02:22:01 | INFO | train rank=0 | step=60550/100000 progress=60.6% loss=0.027817 lr=3.684622e-06 elapsed=22:26:31 avg_step=1.33s recent_step=1.16s eta=14:37:17 +2026-08-26 02:22:16 | INFO | train rank=0 | step=60560/100000 progress=60.6% loss=0.082306 lr=3.683027e-06 elapsed=22:26:46 avg_step=1.33s recent_step=1.52s eta=14:37:05 +2026-08-26 02:22:29 | INFO | train rank=0 | step=60570/100000 progress=60.6% loss=0.045059 lr=3.681432e-06 elapsed=22:26:59 avg_step=1.33s recent_step=1.32s eta=14:36:52 +2026-08-26 02:22:45 | INFO | train rank=0 | step=60580/100000 progress=60.6% loss=0.037613 lr=3.679837e-06 elapsed=22:27:15 avg_step=1.33s recent_step=1.55s eta=14:36:40 +2026-08-26 02:22:56 | INFO | train rank=0 | step=60590/100000 progress=60.6% loss=0.091736 lr=3.678243e-06 elapsed=22:27:26 avg_step=1.33s recent_step=1.13s eta=14:36:25 +2026-08-26 02:23:10 | INFO | train rank=0 | step=60600/100000 progress=60.6% loss=0.078156 lr=3.676648e-06 elapsed=22:27:40 avg_step=1.33s recent_step=1.40s eta=14:36:12 +2026-08-26 02:23:22 | INFO | train rank=0 | step=60610/100000 progress=60.6% loss=0.041260 lr=3.675054e-06 elapsed=22:27:52 avg_step=1.33s recent_step=1.15s eta=14:35:58 +2026-08-26 02:23:36 | INFO | train rank=0 | step=60620/100000 progress=60.6% loss=0.036125 lr=3.673459e-06 elapsed=22:28:06 avg_step=1.33s recent_step=1.40s eta=14:35:45 +2026-08-26 02:23:48 | INFO | train rank=0 | step=60630/100000 progress=60.6% loss=0.045227 lr=3.671865e-06 elapsed=22:28:17 avg_step=1.33s recent_step=1.19s eta=14:35:30 +2026-08-26 02:24:01 | INFO | train rank=0 | step=60640/100000 progress=60.6% loss=0.030441 lr=3.670271e-06 elapsed=22:28:31 avg_step=1.33s recent_step=1.34s eta=14:35:17 +2026-08-26 02:24:15 | INFO | train rank=0 | step=60650/100000 progress=60.7% loss=0.051828 lr=3.668677e-06 elapsed=22:28:45 avg_step=1.33s recent_step=1.44s eta=14:35:05 +2026-08-26 02:24:30 | INFO | train rank=0 | step=60660/100000 progress=60.7% loss=0.041824 lr=3.667084e-06 elapsed=22:29:00 avg_step=1.33s recent_step=1.47s eta=14:34:52 +2026-08-26 02:24:42 | INFO | train rank=0 | step=60670/100000 progress=60.7% loss=0.056580 lr=3.665490e-06 elapsed=22:29:12 avg_step=1.33s recent_step=1.16s eta=14:34:38 +2026-08-26 02:24:54 | INFO | train rank=0 | step=60680/100000 progress=60.7% loss=0.049515 lr=3.663897e-06 elapsed=22:29:24 avg_step=1.33s recent_step=1.27s eta=14:34:24 +2026-08-26 02:25:09 | INFO | train rank=0 | step=60690/100000 progress=60.7% loss=0.050842 lr=3.662303e-06 elapsed=22:29:38 avg_step=1.33s recent_step=1.42s eta=14:34:11 +2026-08-26 02:25:21 | INFO | train rank=0 | step=60700/100000 progress=60.7% loss=0.066772 lr=3.660710e-06 elapsed=22:29:51 avg_step=1.33s recent_step=1.22s eta=14:33:57 +2026-08-26 02:25:37 | INFO | train rank=0 | step=60710/100000 progress=60.7% loss=0.065552 lr=3.659117e-06 elapsed=22:30:07 avg_step=1.33s recent_step=1.66s eta=14:33:46 +2026-08-26 02:25:49 | INFO | train rank=0 | step=60720/100000 progress=60.7% loss=0.063904 lr=3.657524e-06 elapsed=22:30:19 avg_step=1.33s recent_step=1.14s eta=14:33:31 +2026-08-26 02:26:03 | INFO | train rank=0 | step=60730/100000 progress=60.7% loss=0.069763 lr=3.655932e-06 elapsed=22:30:33 avg_step=1.33s recent_step=1.46s eta=14:33:19 +2026-08-26 02:26:15 | INFO | train rank=0 | step=60740/100000 progress=60.7% loss=0.054062 lr=3.654339e-06 elapsed=22:30:45 avg_step=1.33s recent_step=1.17s eta=14:33:04 +2026-08-26 02:26:33 | INFO | train rank=0 | step=60750/100000 progress=60.8% loss=0.070007 lr=3.652747e-06 elapsed=22:31:03 avg_step=1.33s recent_step=1.81s eta=14:32:54 +2026-08-26 02:26:45 | INFO | train rank=0 | step=60760/100000 progress=60.8% loss=0.039673 lr=3.651155e-06 elapsed=22:31:15 avg_step=1.33s recent_step=1.19s eta=14:32:40 +2026-08-26 02:26:58 | INFO | train rank=0 | step=60770/100000 progress=60.8% loss=0.083405 lr=3.649562e-06 elapsed=22:31:28 avg_step=1.33s recent_step=1.29s eta=14:32:26 +2026-08-26 02:27:12 | INFO | train rank=0 | step=60780/100000 progress=60.8% loss=0.032417 lr=3.647971e-06 elapsed=22:31:42 avg_step=1.33s recent_step=1.44s eta=14:32:13 +2026-08-26 02:27:27 | INFO | train rank=0 | step=60790/100000 progress=60.8% loss=0.048065 lr=3.646379e-06 elapsed=22:31:57 avg_step=1.33s recent_step=1.50s eta=14:32:01 +2026-08-26 02:27:39 | INFO | train rank=0 | step=60800/100000 progress=60.8% loss=0.055618 lr=3.644787e-06 elapsed=22:32:08 avg_step=1.33s recent_step=1.12s eta=14:31:46 +2026-08-26 02:27:57 | INFO | train rank=0 | step=60810/100000 progress=60.8% loss=0.035858 lr=3.643196e-06 elapsed=22:32:26 avg_step=1.33s recent_step=1.79s eta=14:31:36 +2026-08-26 02:28:08 | INFO | train rank=0 | step=60820/100000 progress=60.8% loss=0.040161 lr=3.641604e-06 elapsed=22:32:38 avg_step=1.33s recent_step=1.14s eta=14:31:21 +2026-08-26 02:28:23 | INFO | train rank=0 | step=60830/100000 progress=60.8% loss=0.061157 lr=3.640013e-06 elapsed=22:32:53 avg_step=1.33s recent_step=1.52s eta=14:31:09 +2026-08-26 02:28:35 | INFO | train rank=0 | step=60840/100000 progress=60.8% loss=0.027969 lr=3.638422e-06 elapsed=22:33:05 avg_step=1.33s recent_step=1.17s eta=14:30:55 +2026-08-26 02:28:49 | INFO | train rank=0 | step=60850/100000 progress=60.9% loss=0.032730 lr=3.636831e-06 elapsed=22:33:19 avg_step=1.33s recent_step=1.41s eta=14:30:42 +2026-08-26 02:29:01 | INFO | train rank=0 | step=60860/100000 progress=60.9% loss=0.034698 lr=3.635240e-06 elapsed=22:33:31 avg_step=1.33s recent_step=1.24s eta=14:30:28 +2026-08-26 02:29:12 | INFO | train rank=0 | step=60870/100000 progress=60.9% loss=0.060196 lr=3.633650e-06 elapsed=22:33:42 avg_step=1.33s recent_step=1.10s eta=14:30:13 +2026-08-26 02:29:28 | INFO | train rank=0 | step=60880/100000 progress=60.9% loss=0.026062 lr=3.632059e-06 elapsed=22:33:58 avg_step=1.33s recent_step=1.57s eta=14:30:01 +2026-08-26 02:29:40 | INFO | train rank=0 | step=60890/100000 progress=60.9% loss=0.035492 lr=3.630469e-06 elapsed=22:34:10 avg_step=1.33s recent_step=1.19s eta=14:29:47 +2026-08-26 02:29:57 | INFO | train rank=0 | step=60900/100000 progress=60.9% loss=0.057404 lr=3.628879e-06 elapsed=22:34:27 avg_step=1.33s recent_step=1.71s eta=14:29:36 +2026-08-26 02:30:09 | INFO | train rank=0 | step=60910/100000 progress=60.9% loss=0.058510 lr=3.627289e-06 elapsed=22:34:39 avg_step=1.33s recent_step=1.17s eta=14:29:22 +2026-08-26 02:30:24 | INFO | train rank=0 | step=60920/100000 progress=60.9% loss=0.057495 lr=3.625699e-06 elapsed=22:34:53 avg_step=1.33s recent_step=1.49s eta=14:29:09 +2026-08-26 02:30:36 | INFO | train rank=0 | step=60930/100000 progress=60.9% loss=0.040100 lr=3.624109e-06 elapsed=22:35:05 avg_step=1.33s recent_step=1.20s eta=14:28:55 +2026-08-26 02:30:53 | INFO | train rank=0 | step=60940/100000 progress=60.9% loss=0.039108 lr=3.622520e-06 elapsed=22:35:23 avg_step=1.33s recent_step=1.72s eta=14:28:44 +2026-08-26 02:31:06 | INFO | train rank=0 | step=60950/100000 progress=61.0% loss=0.071594 lr=3.620930e-06 elapsed=22:35:35 avg_step=1.33s recent_step=1.28s eta=14:28:31 +2026-08-26 02:31:17 | INFO | train rank=0 | step=60960/100000 progress=61.0% loss=0.031296 lr=3.619341e-06 elapsed=22:35:47 avg_step=1.33s recent_step=1.14s eta=14:28:16 +2026-08-26 02:31:34 | INFO | train rank=0 | step=60970/100000 progress=61.0% loss=0.042389 lr=3.617752e-06 elapsed=22:36:04 avg_step=1.33s recent_step=1.68s eta=14:28:05 +2026-08-26 02:31:45 | INFO | train rank=0 | step=60980/100000 progress=61.0% loss=0.047409 lr=3.616163e-06 elapsed=22:36:15 avg_step=1.33s recent_step=1.11s eta=14:27:50 +2026-08-26 02:31:59 | INFO | train rank=0 | step=60990/100000 progress=61.0% loss=0.042252 lr=3.614574e-06 elapsed=22:36:29 avg_step=1.33s recent_step=1.44s eta=14:27:37 +2026-08-26 02:32:11 | INFO | train rank=0 | step=61000/100000 progress=61.0% loss=0.034286 lr=3.612985e-06 elapsed=22:36:41 avg_step=1.33s recent_step=1.18s eta=14:27:23 +2026-08-26 02:32:23 | INFO | train rank=0 | step=61010/100000 progress=61.0% loss=0.049377 lr=3.611397e-06 elapsed=22:36:52 avg_step=1.33s recent_step=1.15s eta=14:27:09 +2026-08-26 02:32:39 | INFO | train rank=0 | step=61020/100000 progress=61.0% loss=0.030106 lr=3.609809e-06 elapsed=22:37:09 avg_step=1.33s recent_step=1.64s eta=14:26:57 +2026-08-26 02:32:51 | INFO | train rank=0 | step=61030/100000 progress=61.0% loss=0.049278 lr=3.608220e-06 elapsed=22:37:21 avg_step=1.33s recent_step=1.22s eta=14:26:43 +2026-08-26 02:33:05 | INFO | train rank=0 | step=61040/100000 progress=61.0% loss=0.047104 lr=3.606632e-06 elapsed=22:37:34 avg_step=1.33s recent_step=1.34s eta=14:26:30 +2026-08-26 02:33:16 | INFO | train rank=0 | step=61050/100000 progress=61.1% loss=0.039993 lr=3.605044e-06 elapsed=22:37:46 avg_step=1.33s recent_step=1.18s eta=14:26:15 +2026-08-26 02:33:30 | INFO | train rank=0 | step=61060/100000 progress=61.1% loss=0.049438 lr=3.603457e-06 elapsed=22:38:00 avg_step=1.33s recent_step=1.36s eta=14:26:02 +2026-08-26 02:33:45 | INFO | train rank=0 | step=61070/100000 progress=61.1% loss=0.050842 lr=3.601869e-06 elapsed=22:38:15 avg_step=1.33s recent_step=1.52s eta=14:25:50 +2026-08-26 02:33:59 | INFO | train rank=0 | step=61080/100000 progress=61.1% loss=0.024368 lr=3.600282e-06 elapsed=22:38:29 avg_step=1.33s recent_step=1.35s eta=14:25:37 +2026-08-26 02:34:10 | INFO | train rank=0 | step=61090/100000 progress=61.1% loss=0.057724 lr=3.598694e-06 elapsed=22:38:40 avg_step=1.33s recent_step=1.11s eta=14:25:22 +2026-08-26 02:34:22 | INFO | train rank=0 | step=61100/100000 progress=61.1% loss=0.048279 lr=3.597107e-06 elapsed=22:38:52 avg_step=1.33s recent_step=1.21s eta=14:25:08 +2026-08-26 02:34:37 | INFO | train rank=0 | step=61110/100000 progress=61.1% loss=0.046791 lr=3.595520e-06 elapsed=22:39:06 avg_step=1.33s recent_step=1.47s eta=14:24:55 +2026-08-26 02:34:50 | INFO | train rank=0 | step=61120/100000 progress=61.1% loss=0.078430 lr=3.593934e-06 elapsed=22:39:20 avg_step=1.33s recent_step=1.37s eta=14:24:42 +2026-08-26 02:35:02 | INFO | train rank=0 | step=61130/100000 progress=61.1% loss=0.052780 lr=3.592347e-06 elapsed=22:39:32 avg_step=1.33s recent_step=1.14s eta=14:24:28 +2026-08-26 02:35:20 | INFO | train rank=0 | step=61140/100000 progress=61.1% loss=0.054672 lr=3.590760e-06 elapsed=22:39:50 avg_step=1.33s recent_step=1.83s eta=14:24:18 +2026-08-26 02:35:32 | INFO | train rank=0 | step=61150/100000 progress=61.2% loss=0.054993 lr=3.589174e-06 elapsed=22:40:02 avg_step=1.33s recent_step=1.22s eta=14:24:03 +2026-08-26 02:35:47 | INFO | train rank=0 | step=61160/100000 progress=61.2% loss=0.046783 lr=3.587588e-06 elapsed=22:40:17 avg_step=1.33s recent_step=1.52s eta=14:23:51 +2026-08-26 02:36:00 | INFO | train rank=0 | step=61170/100000 progress=61.2% loss=0.038849 lr=3.586002e-06 elapsed=22:40:30 avg_step=1.33s recent_step=1.30s eta=14:23:38 +2026-08-26 02:36:17 | INFO | train rank=0 | step=61180/100000 progress=61.2% loss=0.047932 lr=3.584416e-06 elapsed=22:40:47 avg_step=1.33s recent_step=1.65s eta=14:23:26 +2026-08-26 02:36:28 | INFO | train rank=0 | step=61190/100000 progress=61.2% loss=0.044189 lr=3.582830e-06 elapsed=22:40:58 avg_step=1.33s recent_step=1.15s eta=14:23:12 +2026-08-26 02:36:40 | INFO | train rank=0 | step=61200/100000 progress=61.2% loss=0.060242 lr=3.581244e-06 elapsed=22:41:10 avg_step=1.33s recent_step=1.15s eta=14:22:57 +2026-08-26 02:36:54 | INFO | train rank=0 | step=61210/100000 progress=61.2% loss=0.049149 lr=3.579659e-06 elapsed=22:41:24 avg_step=1.33s recent_step=1.44s eta=14:22:45 +2026-08-26 02:37:11 | INFO | train rank=0 | step=61220/100000 progress=61.2% loss=0.056732 lr=3.578074e-06 elapsed=22:41:41 avg_step=1.33s recent_step=1.65s eta=14:22:33 +2026-08-26 02:37:25 | INFO | train rank=0 | step=61230/100000 progress=61.2% loss=0.061081 lr=3.576489e-06 elapsed=22:41:55 avg_step=1.33s recent_step=1.44s eta=14:22:21 +2026-08-26 02:37:37 | INFO | train rank=0 | step=61240/100000 progress=61.2% loss=0.058044 lr=3.574904e-06 elapsed=22:42:07 avg_step=1.33s recent_step=1.17s eta=14:22:06 +2026-08-26 02:37:51 | INFO | train rank=0 | step=61250/100000 progress=61.3% loss=0.056831 lr=3.573319e-06 elapsed=22:42:21 avg_step=1.33s recent_step=1.41s eta=14:21:53 +2026-08-26 02:38:06 | INFO | train rank=0 | step=61260/100000 progress=61.3% loss=0.045898 lr=3.571734e-06 elapsed=22:42:36 avg_step=1.33s recent_step=1.52s eta=14:21:41 +2026-08-26 02:38:21 | INFO | train rank=0 | step=61270/100000 progress=61.3% loss=0.072906 lr=3.570150e-06 elapsed=22:42:51 avg_step=1.33s recent_step=1.52s eta=14:21:29 +2026-08-26 02:38:34 | INFO | train rank=0 | step=61280/100000 progress=61.3% loss=0.057434 lr=3.568565e-06 elapsed=22:43:04 avg_step=1.33s recent_step=1.23s eta=14:21:15 +2026-08-26 02:38:45 | INFO | train rank=0 | step=61290/100000 progress=61.3% loss=0.048904 lr=3.566981e-06 elapsed=22:43:15 avg_step=1.33s recent_step=1.16s eta=14:21:01 +2026-08-26 02:39:01 | INFO | train rank=0 | step=61300/100000 progress=61.3% loss=0.093384 lr=3.565397e-06 elapsed=22:43:31 avg_step=1.33s recent_step=1.56s eta=14:20:49 +2026-08-26 02:39:13 | INFO | train rank=0 | step=61310/100000 progress=61.3% loss=0.088043 lr=3.563813e-06 elapsed=22:43:43 avg_step=1.33s recent_step=1.22s eta=14:20:35 +2026-08-26 02:39:27 | INFO | train rank=0 | step=61320/100000 progress=61.3% loss=0.038452 lr=3.562230e-06 elapsed=22:43:57 avg_step=1.33s recent_step=1.42s eta=14:20:22 +2026-08-26 02:39:39 | INFO | train rank=0 | step=61330/100000 progress=61.3% loss=0.030968 lr=3.560646e-06 elapsed=22:44:09 avg_step=1.33s recent_step=1.14s eta=14:20:07 +2026-08-26 02:39:54 | INFO | train rank=0 | step=61340/100000 progress=61.3% loss=0.058014 lr=3.559063e-06 elapsed=22:44:24 avg_step=1.33s recent_step=1.54s eta=14:19:55 +2026-08-26 02:40:08 | INFO | train rank=0 | step=61350/100000 progress=61.4% loss=0.076904 lr=3.557479e-06 elapsed=22:44:38 avg_step=1.33s recent_step=1.40s eta=14:19:42 +2026-08-26 02:40:24 | INFO | train rank=0 | step=61360/100000 progress=61.4% loss=0.033188 lr=3.555896e-06 elapsed=22:44:54 avg_step=1.33s recent_step=1.60s eta=14:19:31 +2026-08-26 02:40:37 | INFO | train rank=0 | step=61370/100000 progress=61.4% loss=0.046463 lr=3.554313e-06 elapsed=22:45:07 avg_step=1.33s recent_step=1.30s eta=14:19:17 +2026-08-26 02:40:49 | INFO | train rank=0 | step=61380/100000 progress=61.4% loss=0.032448 lr=3.552731e-06 elapsed=22:45:19 avg_step=1.33s recent_step=1.18s eta=14:19:03 +2026-08-26 02:41:00 | INFO | train rank=0 | step=61390/100000 progress=61.4% loss=0.052696 lr=3.551148e-06 elapsed=22:45:30 avg_step=1.33s recent_step=1.15s eta=14:18:48 +2026-08-26 02:41:13 | INFO | train rank=0 | step=61400/100000 progress=61.4% loss=0.047043 lr=3.549566e-06 elapsed=22:45:43 avg_step=1.33s recent_step=1.31s eta=14:18:35 +2026-08-26 02:41:25 | INFO | train rank=0 | step=61410/100000 progress=61.4% loss=0.039169 lr=3.547983e-06 elapsed=22:45:55 avg_step=1.33s recent_step=1.18s eta=14:18:20 +2026-08-26 02:41:43 | INFO | train rank=0 | step=61420/100000 progress=61.4% loss=0.050484 lr=3.546401e-06 elapsed=22:46:13 avg_step=1.33s recent_step=1.74s eta=14:18:10 +2026-08-26 02:41:54 | INFO | train rank=0 | step=61430/100000 progress=61.4% loss=0.039967 lr=3.544819e-06 elapsed=22:46:24 avg_step=1.33s recent_step=1.15s eta=14:17:55 +2026-08-26 02:42:06 | INFO | train rank=0 | step=61440/100000 progress=61.4% loss=0.052902 lr=3.543237e-06 elapsed=22:46:36 avg_step=1.33s recent_step=1.17s eta=14:17:41 +2026-08-26 02:42:18 | INFO | train rank=0 | step=61450/100000 progress=61.5% loss=0.045120 lr=3.541656e-06 elapsed=22:46:48 avg_step=1.33s recent_step=1.20s eta=14:17:26 +2026-08-26 02:42:33 | INFO | train rank=0 | step=61460/100000 progress=61.5% loss=0.061508 lr=3.540074e-06 elapsed=22:47:03 avg_step=1.33s recent_step=1.50s eta=14:17:14 +2026-08-26 02:42:44 | INFO | train rank=0 | step=61470/100000 progress=61.5% loss=0.075775 lr=3.538493e-06 elapsed=22:47:14 avg_step=1.33s recent_step=1.16s eta=14:17:00 +2026-08-26 02:42:58 | INFO | train rank=0 | step=61480/100000 progress=61.5% loss=0.044952 lr=3.536912e-06 elapsed=22:47:28 avg_step=1.33s recent_step=1.37s eta=14:16:47 +2026-08-26 02:43:13 | INFO | train rank=0 | step=61490/100000 progress=61.5% loss=0.051743 lr=3.535331e-06 elapsed=22:47:42 avg_step=1.33s recent_step=1.44s eta=14:16:34 +2026-08-26 02:43:27 | INFO | train rank=0 | step=61500/100000 progress=61.5% loss=0.069641 lr=3.533750e-06 elapsed=22:47:57 avg_step=1.33s recent_step=1.44s eta=14:16:21 +2026-08-26 02:43:41 | INFO | train rank=0 | step=61510/100000 progress=61.5% loss=0.038895 lr=3.532169e-06 elapsed=22:48:11 avg_step=1.33s recent_step=1.43s eta=14:16:09 +2026-08-26 02:43:57 | INFO | train rank=0 | step=61520/100000 progress=61.5% loss=0.092529 lr=3.530589e-06 elapsed=22:48:27 avg_step=1.33s recent_step=1.55s eta=14:15:57 +2026-08-26 02:44:08 | INFO | train rank=0 | step=61530/100000 progress=61.5% loss=0.065277 lr=3.529008e-06 elapsed=22:48:38 avg_step=1.33s recent_step=1.15s eta=14:15:42 +2026-08-26 02:44:23 | INFO | train rank=0 | step=61540/100000 progress=61.5% loss=0.053078 lr=3.527428e-06 elapsed=22:48:52 avg_step=1.33s recent_step=1.43s eta=14:15:29 +2026-08-26 02:44:34 | INFO | train rank=0 | step=61550/100000 progress=61.6% loss=0.032135 lr=3.525848e-06 elapsed=22:49:04 avg_step=1.33s recent_step=1.18s eta=14:15:15 +2026-08-26 02:44:54 | INFO | train rank=0 | step=61560/100000 progress=61.6% loss=0.076782 lr=3.524268e-06 elapsed=22:49:24 avg_step=1.33s recent_step=1.98s eta=14:15:06 +2026-08-26 02:45:06 | INFO | train rank=0 | step=61570/100000 progress=61.6% loss=0.080566 lr=3.522688e-06 elapsed=22:49:36 avg_step=1.33s recent_step=1.16s eta=14:14:51 +2026-08-26 02:45:17 | INFO | train rank=0 | step=61580/100000 progress=61.6% loss=0.065155 lr=3.521109e-06 elapsed=22:49:47 avg_step=1.33s recent_step=1.14s eta=14:14:37 +2026-08-26 02:45:29 | INFO | train rank=0 | step=61590/100000 progress=61.6% loss=0.062210 lr=3.519529e-06 elapsed=22:49:59 avg_step=1.33s recent_step=1.21s eta=14:14:23 +2026-08-26 02:45:43 | INFO | train rank=0 | step=61600/100000 progress=61.6% loss=0.078094 lr=3.517950e-06 elapsed=22:50:13 avg_step=1.33s recent_step=1.41s eta=14:14:10 +2026-08-26 02:45:55 | INFO | train rank=0 | step=61610/100000 progress=61.6% loss=0.044785 lr=3.516371e-06 elapsed=22:50:25 avg_step=1.33s recent_step=1.16s eta=14:13:55 +2026-08-26 02:46:10 | INFO | train rank=0 | step=61620/100000 progress=61.6% loss=0.036835 lr=3.514792e-06 elapsed=22:50:40 avg_step=1.33s recent_step=1.48s eta=14:13:43 +2026-08-26 02:46:22 | INFO | train rank=0 | step=61630/100000 progress=61.6% loss=0.057281 lr=3.513213e-06 elapsed=22:50:52 avg_step=1.33s recent_step=1.18s eta=14:13:29 +2026-08-26 02:46:36 | INFO | train rank=0 | step=61640/100000 progress=61.6% loss=0.040848 lr=3.511635e-06 elapsed=22:51:06 avg_step=1.33s recent_step=1.44s eta=14:13:16 +2026-08-26 02:46:50 | INFO | train rank=0 | step=61650/100000 progress=61.7% loss=0.037071 lr=3.510056e-06 elapsed=22:51:20 avg_step=1.33s recent_step=1.41s eta=14:13:03 +2026-08-26 02:47:07 | INFO | train rank=0 | step=61660/100000 progress=61.7% loss=0.034889 lr=3.508478e-06 elapsed=22:51:37 avg_step=1.33s recent_step=1.67s eta=14:12:52 +2026-08-26 02:47:18 | INFO | train rank=0 | step=61670/100000 progress=61.7% loss=0.055908 lr=3.506900e-06 elapsed=22:51:48 avg_step=1.33s recent_step=1.15s eta=14:12:37 +2026-08-26 02:47:32 | INFO | train rank=0 | step=61680/100000 progress=61.7% loss=0.071594 lr=3.505322e-06 elapsed=22:52:02 avg_step=1.33s recent_step=1.39s eta=14:12:24 +2026-08-26 02:47:46 | INFO | train rank=0 | step=61690/100000 progress=61.7% loss=0.066162 lr=3.503744e-06 elapsed=22:52:15 avg_step=1.33s recent_step=1.32s eta=14:12:11 +2026-08-26 02:48:03 | INFO | train rank=0 | step=61700/100000 progress=61.7% loss=0.039429 lr=3.502167e-06 elapsed=22:52:32 avg_step=1.33s recent_step=1.70s eta=14:12:00 +2026-08-26 02:48:14 | INFO | train rank=0 | step=61710/100000 progress=61.7% loss=0.033081 lr=3.500589e-06 elapsed=22:52:44 avg_step=1.33s recent_step=1.18s eta=14:11:45 +2026-08-26 02:48:26 | INFO | train rank=0 | step=61720/100000 progress=61.7% loss=0.050446 lr=3.499012e-06 elapsed=22:52:56 avg_step=1.33s recent_step=1.19s eta=14:11:31 +2026-08-26 02:48:38 | INFO | train rank=0 | step=61730/100000 progress=61.7% loss=0.026085 lr=3.497435e-06 elapsed=22:53:08 avg_step=1.33s recent_step=1.18s eta=14:11:17 +2026-08-26 02:48:53 | INFO | train rank=0 | step=61740/100000 progress=61.7% loss=0.053467 lr=3.495858e-06 elapsed=22:53:23 avg_step=1.33s recent_step=1.49s eta=14:11:05 +2026-08-26 02:49:08 | INFO | train rank=0 | step=61750/100000 progress=61.8% loss=0.031502 lr=3.494281e-06 elapsed=22:53:38 avg_step=1.33s recent_step=1.50s eta=14:10:52 +2026-08-26 02:49:22 | INFO | train rank=0 | step=61760/100000 progress=61.8% loss=0.038635 lr=3.492704e-06 elapsed=22:53:51 avg_step=1.33s recent_step=1.34s eta=14:10:39 +2026-08-26 02:49:33 | INFO | train rank=0 | step=61770/100000 progress=61.8% loss=0.036705 lr=3.491128e-06 elapsed=22:54:03 avg_step=1.33s recent_step=1.19s eta=14:10:25 +2026-08-26 02:49:49 | INFO | train rank=0 | step=61780/100000 progress=61.8% loss=0.069580 lr=3.489552e-06 elapsed=22:54:19 avg_step=1.33s recent_step=1.60s eta=14:10:13 +2026-08-26 02:50:01 | INFO | train rank=0 | step=61790/100000 progress=61.8% loss=0.045044 lr=3.487975e-06 elapsed=22:54:31 avg_step=1.33s recent_step=1.16s eta=14:09:59 +2026-08-26 02:50:16 | INFO | train rank=0 | step=61800/100000 progress=61.8% loss=0.035080 lr=3.486399e-06 elapsed=22:54:46 avg_step=1.33s recent_step=1.50s eta=14:09:46 +2026-08-26 02:50:28 | INFO | train rank=0 | step=61810/100000 progress=61.8% loss=0.048645 lr=3.484824e-06 elapsed=22:54:58 avg_step=1.33s recent_step=1.18s eta=14:09:32 +2026-08-26 02:50:41 | INFO | train rank=0 | step=61820/100000 progress=61.8% loss=0.059784 lr=3.483248e-06 elapsed=22:55:10 avg_step=1.33s recent_step=1.28s eta=14:09:18 +2026-08-26 02:50:52 | INFO | train rank=0 | step=61830/100000 progress=61.8% loss=0.052689 lr=3.481673e-06 elapsed=22:55:22 avg_step=1.33s recent_step=1.14s eta=14:09:04 +2026-08-26 02:51:08 | INFO | train rank=0 | step=61840/100000 progress=61.8% loss=0.046181 lr=3.480097e-06 elapsed=22:55:38 avg_step=1.33s recent_step=1.58s eta=14:08:52 +2026-08-26 02:51:19 | INFO | train rank=0 | step=61850/100000 progress=61.9% loss=0.035019 lr=3.478522e-06 elapsed=22:55:49 avg_step=1.33s recent_step=1.09s eta=14:08:37 +2026-08-26 02:51:31 | INFO | train rank=0 | step=61860/100000 progress=61.9% loss=0.045898 lr=3.476947e-06 elapsed=22:56:01 avg_step=1.33s recent_step=1.21s eta=14:08:23 +2026-08-26 02:51:47 | INFO | train rank=0 | step=61870/100000 progress=61.9% loss=0.046997 lr=3.475372e-06 elapsed=22:56:17 avg_step=1.33s recent_step=1.59s eta=14:08:11 +2026-08-26 02:51:59 | INFO | train rank=0 | step=61880/100000 progress=61.9% loss=0.041351 lr=3.473798e-06 elapsed=22:56:29 avg_step=1.33s recent_step=1.22s eta=14:07:57 +2026-08-26 02:52:13 | INFO | train rank=0 | step=61890/100000 progress=61.9% loss=0.069687 lr=3.472223e-06 elapsed=22:56:43 avg_step=1.33s recent_step=1.45s eta=14:07:44 +2026-08-26 02:52:27 | INFO | train rank=0 | step=61900/100000 progress=61.9% loss=0.056976 lr=3.470649e-06 elapsed=22:56:57 avg_step=1.33s recent_step=1.36s eta=14:07:31 +2026-08-26 02:52:40 | INFO | train rank=0 | step=61910/100000 progress=61.9% loss=0.056580 lr=3.469075e-06 elapsed=22:57:10 avg_step=1.33s recent_step=1.33s eta=14:07:18 +2026-08-26 02:52:52 | INFO | train rank=0 | step=61920/100000 progress=61.9% loss=0.071594 lr=3.467501e-06 elapsed=22:57:22 avg_step=1.33s recent_step=1.15s eta=14:07:03 +2026-08-26 02:53:06 | INFO | train rank=0 | step=61930/100000 progress=61.9% loss=0.025928 lr=3.465927e-06 elapsed=22:57:36 avg_step=1.33s recent_step=1.39s eta=14:06:50 +2026-08-26 02:53:19 | INFO | train rank=0 | step=61940/100000 progress=61.9% loss=0.064728 lr=3.464353e-06 elapsed=22:57:49 avg_step=1.33s recent_step=1.36s eta=14:06:37 +2026-08-26 02:53:31 | INFO | train rank=0 | step=61950/100000 progress=62.0% loss=0.034683 lr=3.462780e-06 elapsed=22:58:01 avg_step=1.33s recent_step=1.18s eta=14:06:23 +2026-08-26 02:53:42 | INFO | train rank=0 | step=61960/100000 progress=62.0% loss=0.052887 lr=3.461207e-06 elapsed=22:58:12 avg_step=1.33s recent_step=1.09s eta=14:06:08 +2026-08-26 02:53:54 | INFO | train rank=0 | step=61970/100000 progress=62.0% loss=0.058014 lr=3.459633e-06 elapsed=22:58:24 avg_step=1.33s recent_step=1.19s eta=14:05:54 +2026-08-26 02:54:05 | INFO | train rank=0 | step=61980/100000 progress=62.0% loss=0.035744 lr=3.458060e-06 elapsed=22:58:35 avg_step=1.33s recent_step=1.12s eta=14:05:39 +2026-08-26 02:54:17 | INFO | train rank=0 | step=61990/100000 progress=62.0% loss=0.048592 lr=3.456488e-06 elapsed=22:58:47 avg_step=1.33s recent_step=1.18s eta=14:05:25 +2026-08-26 02:54:34 | INFO | train rank=0 | step=62000/100000 progress=62.0% loss=0.040924 lr=3.454915e-06 elapsed=22:59:04 avg_step=1.33s recent_step=1.70s eta=14:05:14 +2026-08-26 02:54:46 | INFO | train rank=0 | step=62010/100000 progress=62.0% loss=0.049957 lr=3.453343e-06 elapsed=22:59:16 avg_step=1.33s recent_step=1.19s eta=14:05:00 +2026-08-26 02:54:58 | INFO | train rank=0 | step=62020/100000 progress=62.0% loss=0.031075 lr=3.451770e-06 elapsed=22:59:28 avg_step=1.33s recent_step=1.21s eta=14:04:45 +2026-08-26 02:55:12 | INFO | train rank=0 | step=62030/100000 progress=62.0% loss=0.045624 lr=3.450198e-06 elapsed=22:59:42 avg_step=1.33s recent_step=1.44s eta=14:04:33 +2026-08-26 02:55:25 | INFO | train rank=0 | step=62040/100000 progress=62.0% loss=0.032936 lr=3.448626e-06 elapsed=22:59:55 avg_step=1.33s recent_step=1.29s eta=14:04:19 +2026-08-26 02:55:39 | INFO | train rank=0 | step=62050/100000 progress=62.1% loss=0.036713 lr=3.447054e-06 elapsed=23:00:09 avg_step=1.33s recent_step=1.37s eta=14:04:06 +2026-08-26 02:55:56 | INFO | train rank=0 | step=62060/100000 progress=62.1% loss=0.019180 lr=3.445483e-06 elapsed=23:00:25 avg_step=1.33s recent_step=1.65s eta=14:03:55 +2026-08-26 02:56:07 | INFO | train rank=0 | step=62070/100000 progress=62.1% loss=0.072113 lr=3.443911e-06 elapsed=23:00:37 avg_step=1.33s recent_step=1.17s eta=14:03:40 +2026-08-26 02:56:20 | INFO | train rank=0 | step=62080/100000 progress=62.1% loss=0.021538 lr=3.442340e-06 elapsed=23:00:50 avg_step=1.33s recent_step=1.32s eta=14:03:27 +2026-08-26 02:56:34 | INFO | train rank=0 | step=62090/100000 progress=62.1% loss=0.046494 lr=3.440769e-06 elapsed=23:01:04 avg_step=1.33s recent_step=1.36s eta=14:03:14 +2026-08-26 02:56:47 | INFO | train rank=0 | step=62100/100000 progress=62.1% loss=0.041222 lr=3.439198e-06 elapsed=23:01:17 avg_step=1.33s recent_step=1.31s eta=14:03:00 +2026-08-26 02:56:59 | INFO | train rank=0 | step=62110/100000 progress=62.1% loss=0.045044 lr=3.437627e-06 elapsed=23:01:29 avg_step=1.33s recent_step=1.18s eta=14:02:46 +2026-08-26 02:57:13 | INFO | train rank=0 | step=62120/100000 progress=62.1% loss=0.036102 lr=3.436057e-06 elapsed=23:01:43 avg_step=1.33s recent_step=1.41s eta=14:02:33 +2026-08-26 02:57:31 | INFO | train rank=0 | step=62130/100000 progress=62.1% loss=0.040939 lr=3.434486e-06 elapsed=23:02:01 avg_step=1.33s recent_step=1.75s eta=14:02:22 +2026-08-26 02:57:42 | INFO | train rank=0 | step=62140/100000 progress=62.1% loss=0.031517 lr=3.432916e-06 elapsed=23:02:12 avg_step=1.33s recent_step=1.10s eta=14:02:07 +2026-08-26 02:57:58 | INFO | train rank=0 | step=62150/100000 progress=62.2% loss=0.047516 lr=3.431346e-06 elapsed=23:02:28 avg_step=1.33s recent_step=1.68s eta=14:01:56 +2026-08-26 02:58:10 | INFO | train rank=0 | step=62160/100000 progress=62.2% loss=0.047684 lr=3.429776e-06 elapsed=23:02:40 avg_step=1.33s recent_step=1.19s eta=14:01:42 +2026-08-26 02:58:23 | INFO | train rank=0 | step=62170/100000 progress=62.2% loss=0.050430 lr=3.428206e-06 elapsed=23:02:53 avg_step=1.33s recent_step=1.23s eta=14:01:28 +2026-08-26 02:58:42 | INFO | train rank=0 | step=62180/100000 progress=62.2% loss=0.064148 lr=3.426637e-06 elapsed=23:03:12 avg_step=1.33s recent_step=1.92s eta=14:01:18 +2026-08-26 02:58:53 | INFO | train rank=0 | step=62190/100000 progress=62.2% loss=0.072510 lr=3.425067e-06 elapsed=23:03:23 avg_step=1.33s recent_step=1.13s eta=14:01:04 +2026-08-26 02:59:07 | INFO | train rank=0 | step=62200/100000 progress=62.2% loss=0.044617 lr=3.423498e-06 elapsed=23:03:37 avg_step=1.33s recent_step=1.43s eta=14:00:51 +2026-08-26 02:59:19 | INFO | train rank=0 | step=62210/100000 progress=62.2% loss=0.043930 lr=3.421929e-06 elapsed=23:03:49 avg_step=1.33s recent_step=1.20s eta=14:00:37 +2026-08-26 02:59:35 | INFO | train rank=0 | step=62220/100000 progress=62.2% loss=0.071777 lr=3.420360e-06 elapsed=23:04:05 avg_step=1.33s recent_step=1.58s eta=14:00:25 +2026-08-26 02:59:50 | INFO | train rank=0 | step=62230/100000 progress=62.2% loss=0.025070 lr=3.418792e-06 elapsed=23:04:20 avg_step=1.33s recent_step=1.49s eta=14:00:12 +2026-08-26 03:00:03 | INFO | train rank=0 | step=62240/100000 progress=62.2% loss=0.065689 lr=3.417223e-06 elapsed=23:04:33 avg_step=1.33s recent_step=1.29s eta=13:59:59 +2026-08-26 03:00:19 | INFO | train rank=0 | step=62250/100000 progress=62.3% loss=0.045479 lr=3.415655e-06 elapsed=23:04:49 avg_step=1.33s recent_step=1.62s eta=13:59:47 +2026-08-26 03:00:35 | INFO | train rank=0 | step=62260/100000 progress=62.3% loss=0.048157 lr=3.414087e-06 elapsed=23:05:05 avg_step=1.33s recent_step=1.56s eta=13:59:35 +2026-08-26 03:00:46 | INFO | train rank=0 | step=62270/100000 progress=62.3% loss=0.076904 lr=3.412519e-06 elapsed=23:05:16 avg_step=1.33s recent_step=1.11s eta=13:59:20 +2026-08-26 03:00:59 | INFO | train rank=0 | step=62280/100000 progress=62.3% loss=0.055252 lr=3.410951e-06 elapsed=23:05:29 avg_step=1.33s recent_step=1.30s eta=13:59:07 +2026-08-26 03:01:13 | INFO | train rank=0 | step=62290/100000 progress=62.3% loss=0.040039 lr=3.409383e-06 elapsed=23:05:43 avg_step=1.33s recent_step=1.42s eta=13:58:54 +2026-08-26 03:01:24 | INFO | train rank=0 | step=62300/100000 progress=62.3% loss=0.054596 lr=3.407816e-06 elapsed=23:05:54 avg_step=1.33s recent_step=1.12s eta=13:58:39 +2026-08-26 03:01:38 | INFO | train rank=0 | step=62310/100000 progress=62.3% loss=0.044418 lr=3.406248e-06 elapsed=23:06:08 avg_step=1.33s recent_step=1.38s eta=13:58:26 +2026-08-26 03:01:50 | INFO | train rank=0 | step=62320/100000 progress=62.3% loss=0.028481 lr=3.404681e-06 elapsed=23:06:20 avg_step=1.33s recent_step=1.17s eta=13:58:12 +2026-08-26 03:02:04 | INFO | train rank=0 | step=62330/100000 progress=62.3% loss=0.156372 lr=3.403114e-06 elapsed=23:06:33 avg_step=1.33s recent_step=1.38s eta=13:57:59 +2026-08-26 03:02:18 | INFO | train rank=0 | step=62340/100000 progress=62.3% loss=0.034615 lr=3.401547e-06 elapsed=23:06:48 avg_step=1.33s recent_step=1.43s eta=13:57:46 +2026-08-26 03:02:35 | INFO | train rank=0 | step=62350/100000 progress=62.4% loss=0.053223 lr=3.399981e-06 elapsed=23:07:05 avg_step=1.33s recent_step=1.70s eta=13:57:35 +2026-08-26 03:02:46 | INFO | train rank=0 | step=62360/100000 progress=62.4% loss=0.057983 lr=3.398414e-06 elapsed=23:07:16 avg_step=1.33s recent_step=1.11s eta=13:57:20 +2026-08-26 03:02:58 | INFO | train rank=0 | step=62370/100000 progress=62.4% loss=0.075256 lr=3.396848e-06 elapsed=23:07:28 avg_step=1.33s recent_step=1.18s eta=13:57:06 +2026-08-26 03:03:11 | INFO | train rank=0 | step=62380/100000 progress=62.4% loss=0.033180 lr=3.395282e-06 elapsed=23:07:41 avg_step=1.33s recent_step=1.32s eta=13:56:53 +2026-08-26 03:03:24 | INFO | train rank=0 | step=62390/100000 progress=62.4% loss=0.048141 lr=3.393716e-06 elapsed=23:07:54 avg_step=1.33s recent_step=1.30s eta=13:56:39 +2026-08-26 03:03:37 | INFO | train rank=0 | step=62400/100000 progress=62.4% loss=0.065704 lr=3.392150e-06 elapsed=23:08:07 avg_step=1.33s recent_step=1.34s eta=13:56:26 +2026-08-26 03:03:49 | INFO | train rank=0 | step=62410/100000 progress=62.4% loss=0.038773 lr=3.390585e-06 elapsed=23:08:19 avg_step=1.33s recent_step=1.13s eta=13:56:11 +2026-08-26 03:04:03 | INFO | train rank=0 | step=62420/100000 progress=62.4% loss=0.020386 lr=3.389019e-06 elapsed=23:08:32 avg_step=1.33s recent_step=1.38s eta=13:55:58 +2026-08-26 03:04:17 | INFO | train rank=0 | step=62430/100000 progress=62.4% loss=0.063797 lr=3.387454e-06 elapsed=23:08:47 avg_step=1.33s recent_step=1.46s eta=13:55:46 +2026-08-26 03:04:30 | INFO | train rank=0 | step=62440/100000 progress=62.4% loss=0.050934 lr=3.385889e-06 elapsed=23:09:00 avg_step=1.33s recent_step=1.28s eta=13:55:32 +2026-08-26 03:04:44 | INFO | train rank=0 | step=62450/100000 progress=62.5% loss=0.056412 lr=3.384324e-06 elapsed=23:09:14 avg_step=1.33s recent_step=1.43s eta=13:55:19 +2026-08-26 03:04:56 | INFO | train rank=0 | step=62460/100000 progress=62.5% loss=0.054382 lr=3.382760e-06 elapsed=23:09:26 avg_step=1.33s recent_step=1.16s eta=13:55:05 +2026-08-26 03:05:09 | INFO | train rank=0 | step=62470/100000 progress=62.5% loss=0.064362 lr=3.381195e-06 elapsed=23:09:39 avg_step=1.33s recent_step=1.36s eta=13:54:51 +2026-08-26 03:05:27 | INFO | train rank=0 | step=62480/100000 progress=62.5% loss=0.052597 lr=3.379631e-06 elapsed=23:09:57 avg_step=1.33s recent_step=1.78s eta=13:54:41 +2026-08-26 03:05:39 | INFO | train rank=0 | step=62490/100000 progress=62.5% loss=0.061417 lr=3.378067e-06 elapsed=23:10:09 avg_step=1.33s recent_step=1.15s eta=13:54:26 +2026-08-26 03:05:50 | INFO | train rank=0 | step=62500/100000 progress=62.5% loss=0.057251 lr=3.376503e-06 elapsed=23:10:20 avg_step=1.33s recent_step=1.18s eta=13:54:12 +2026-08-26 03:06:02 | INFO | train rank=0 | step=62510/100000 progress=62.5% loss=0.082153 lr=3.374939e-06 elapsed=23:10:32 avg_step=1.33s recent_step=1.13s eta=13:53:57 +2026-08-26 03:06:15 | INFO | train rank=0 | step=62520/100000 progress=62.5% loss=0.085876 lr=3.373375e-06 elapsed=23:10:45 avg_step=1.33s recent_step=1.31s eta=13:53:44 +2026-08-26 03:06:34 | INFO | train rank=0 | step=62530/100000 progress=62.5% loss=0.065033 lr=3.371812e-06 elapsed=23:11:04 avg_step=1.33s recent_step=1.92s eta=13:53:34 +2026-08-26 03:06:49 | INFO | train rank=0 | step=62540/100000 progress=62.5% loss=0.043503 lr=3.370249e-06 elapsed=23:11:19 avg_step=1.33s recent_step=1.51s eta=13:53:22 +2026-08-26 03:07:01 | INFO | train rank=0 | step=62550/100000 progress=62.5% loss=0.040634 lr=3.368685e-06 elapsed=23:11:31 avg_step=1.33s recent_step=1.21s eta=13:53:08 +2026-08-26 03:07:16 | INFO | train rank=0 | step=62560/100000 progress=62.6% loss=0.030807 lr=3.367123e-06 elapsed=23:11:45 avg_step=1.33s recent_step=1.43s eta=13:52:55 +2026-08-26 03:07:27 | INFO | train rank=0 | step=62570/100000 progress=62.6% loss=0.053680 lr=3.365560e-06 elapsed=23:11:57 avg_step=1.33s recent_step=1.14s eta=13:52:40 +2026-08-26 03:07:42 | INFO | train rank=0 | step=62580/100000 progress=62.6% loss=0.042610 lr=3.363997e-06 elapsed=23:12:12 avg_step=1.33s recent_step=1.51s eta=13:52:28 +2026-08-26 03:07:54 | INFO | train rank=0 | step=62590/100000 progress=62.6% loss=0.049759 lr=3.362435e-06 elapsed=23:12:24 avg_step=1.33s recent_step=1.23s eta=13:52:14 +2026-08-26 03:08:09 | INFO | train rank=0 | step=62600/100000 progress=62.6% loss=0.044540 lr=3.360873e-06 elapsed=23:12:39 avg_step=1.33s recent_step=1.49s eta=13:52:02 +2026-08-26 03:08:21 | INFO | train rank=0 | step=62610/100000 progress=62.6% loss=0.037697 lr=3.359311e-06 elapsed=23:12:51 avg_step=1.33s recent_step=1.15s eta=13:51:47 +2026-08-26 03:08:36 | INFO | train rank=0 | step=62620/100000 progress=62.6% loss=0.044678 lr=3.357749e-06 elapsed=23:13:06 avg_step=1.33s recent_step=1.52s eta=13:51:35 +2026-08-26 03:08:48 | INFO | train rank=0 | step=62630/100000 progress=62.6% loss=0.067413 lr=3.356187e-06 elapsed=23:13:18 avg_step=1.33s recent_step=1.17s eta=13:51:21 +2026-08-26 03:09:06 | INFO | train rank=0 | step=62640/100000 progress=62.6% loss=0.063995 lr=3.354626e-06 elapsed=23:13:36 avg_step=1.33s recent_step=1.85s eta=13:51:10 +2026-08-26 03:09:21 | INFO | train rank=0 | step=62650/100000 progress=62.6% loss=0.060211 lr=3.353065e-06 elapsed=23:13:50 avg_step=1.33s recent_step=1.44s eta=13:50:58 +2026-08-26 03:09:32 | INFO | train rank=0 | step=62660/100000 progress=62.7% loss=0.043091 lr=3.351503e-06 elapsed=23:14:02 avg_step=1.33s recent_step=1.17s eta=13:50:43 +2026-08-26 03:09:48 | INFO | train rank=0 | step=62670/100000 progress=62.7% loss=0.038177 lr=3.349942e-06 elapsed=23:14:18 avg_step=1.33s recent_step=1.54s eta=13:50:31 +2026-08-26 03:10:01 | INFO | train rank=0 | step=62680/100000 progress=62.7% loss=0.044525 lr=3.348382e-06 elapsed=23:14:31 avg_step=1.33s recent_step=1.34s eta=13:50:18 +2026-08-26 03:10:13 | INFO | train rank=0 | step=62690/100000 progress=62.7% loss=0.080994 lr=3.346821e-06 elapsed=23:14:42 avg_step=1.33s recent_step=1.15s eta=13:50:03 +2026-08-26 03:10:25 | INFO | train rank=0 | step=62700/100000 progress=62.7% loss=0.038086 lr=3.345261e-06 elapsed=23:14:55 avg_step=1.33s recent_step=1.28s eta=13:49:50 +2026-08-26 03:10:37 | INFO | train rank=0 | step=62710/100000 progress=62.7% loss=0.044067 lr=3.343701e-06 elapsed=23:15:07 avg_step=1.33s recent_step=1.15s eta=13:49:35 +2026-08-26 03:10:52 | INFO | train rank=0 | step=62720/100000 progress=62.7% loss=0.050720 lr=3.342141e-06 elapsed=23:15:22 avg_step=1.33s recent_step=1.55s eta=13:49:23 +2026-08-26 03:11:04 | INFO | train rank=0 | step=62730/100000 progress=62.7% loss=0.039200 lr=3.340581e-06 elapsed=23:15:34 avg_step=1.33s recent_step=1.18s eta=13:49:09 +2026-08-26 03:11:16 | INFO | train rank=0 | step=62740/100000 progress=62.7% loss=0.032196 lr=3.339021e-06 elapsed=23:15:46 avg_step=1.33s recent_step=1.14s eta=13:48:55 +2026-08-26 03:11:31 | INFO | train rank=0 | step=62750/100000 progress=62.7% loss=0.033966 lr=3.337462e-06 elapsed=23:16:01 avg_step=1.33s recent_step=1.51s eta=13:48:42 +2026-08-26 03:11:46 | INFO | train rank=0 | step=62760/100000 progress=62.8% loss=0.046211 lr=3.335902e-06 elapsed=23:16:16 avg_step=1.33s recent_step=1.53s eta=13:48:30 +2026-08-26 03:11:58 | INFO | train rank=0 | step=62770/100000 progress=62.8% loss=0.045967 lr=3.334343e-06 elapsed=23:16:27 avg_step=1.33s recent_step=1.15s eta=13:48:16 +2026-08-26 03:12:11 | INFO | train rank=0 | step=62780/100000 progress=62.8% loss=0.052368 lr=3.332784e-06 elapsed=23:16:41 avg_step=1.33s recent_step=1.39s eta=13:48:03 +2026-08-26 03:12:23 | INFO | train rank=0 | step=62790/100000 progress=62.8% loss=0.094147 lr=3.331225e-06 elapsed=23:16:53 avg_step=1.33s recent_step=1.13s eta=13:47:48 +2026-08-26 03:12:39 | INFO | train rank=0 | step=62800/100000 progress=62.8% loss=0.056427 lr=3.329667e-06 elapsed=23:17:08 avg_step=1.33s recent_step=1.58s eta=13:47:36 +2026-08-26 03:12:51 | INFO | train rank=0 | step=62810/100000 progress=62.8% loss=0.046982 lr=3.328109e-06 elapsed=23:17:21 avg_step=1.33s recent_step=1.25s eta=13:47:22 +2026-08-26 03:13:04 | INFO | train rank=0 | step=62820/100000 progress=62.8% loss=0.042252 lr=3.326550e-06 elapsed=23:17:34 avg_step=1.33s recent_step=1.33s eta=13:47:09 +2026-08-26 03:13:16 | INFO | train rank=0 | step=62830/100000 progress=62.8% loss=0.048855 lr=3.324992e-06 elapsed=23:17:46 avg_step=1.33s recent_step=1.17s eta=13:46:55 +2026-08-26 03:13:34 | INFO | train rank=0 | step=62840/100000 progress=62.8% loss=0.042267 lr=3.323434e-06 elapsed=23:18:04 avg_step=1.33s recent_step=1.83s eta=13:46:44 +2026-08-26 03:13:46 | INFO | train rank=0 | step=62850/100000 progress=62.8% loss=0.041214 lr=3.321877e-06 elapsed=23:18:16 avg_step=1.33s recent_step=1.17s eta=13:46:30 +2026-08-26 03:14:00 | INFO | train rank=0 | step=62860/100000 progress=62.9% loss=0.061127 lr=3.320319e-06 elapsed=23:18:30 avg_step=1.33s recent_step=1.36s eta=13:46:17 +2026-08-26 03:14:14 | INFO | train rank=0 | step=62870/100000 progress=62.9% loss=0.026772 lr=3.318762e-06 elapsed=23:18:44 avg_step=1.33s recent_step=1.45s eta=13:46:04 +2026-08-26 03:14:26 | INFO | train rank=0 | step=62880/100000 progress=62.9% loss=0.055130 lr=3.317205e-06 elapsed=23:18:56 avg_step=1.33s recent_step=1.22s eta=13:45:50 +2026-08-26 03:14:44 | INFO | train rank=0 | step=62890/100000 progress=62.9% loss=0.054199 lr=3.315648e-06 elapsed=23:19:14 avg_step=1.33s recent_step=1.78s eta=13:45:39 +2026-08-26 03:14:55 | INFO | train rank=0 | step=62900/100000 progress=62.9% loss=0.050735 lr=3.314091e-06 elapsed=23:19:25 avg_step=1.33s recent_step=1.11s eta=13:45:24 +2026-08-26 03:15:09 | INFO | train rank=0 | step=62910/100000 progress=62.9% loss=0.029488 lr=3.312535e-06 elapsed=23:19:39 avg_step=1.33s recent_step=1.40s eta=13:45:11 +2026-08-26 03:15:23 | INFO | train rank=0 | step=62920/100000 progress=62.9% loss=0.046280 lr=3.310978e-06 elapsed=23:19:53 avg_step=1.33s recent_step=1.39s eta=13:44:58 +2026-08-26 03:15:35 | INFO | train rank=0 | step=62930/100000 progress=62.9% loss=0.055496 lr=3.309422e-06 elapsed=23:20:05 avg_step=1.33s recent_step=1.17s eta=13:44:44 +2026-08-26 03:15:48 | INFO | train rank=0 | step=62940/100000 progress=62.9% loss=0.027962 lr=3.307866e-06 elapsed=23:20:18 avg_step=1.33s recent_step=1.34s eta=13:44:31 +2026-08-26 03:16:00 | INFO | train rank=0 | step=62950/100000 progress=62.9% loss=0.055267 lr=3.306310e-06 elapsed=23:20:29 avg_step=1.33s recent_step=1.15s eta=13:44:16 +2026-08-26 03:16:11 | INFO | train rank=0 | step=62960/100000 progress=63.0% loss=0.050995 lr=3.304755e-06 elapsed=23:20:41 avg_step=1.33s recent_step=1.15s eta=13:44:02 +2026-08-26 03:16:25 | INFO | train rank=0 | step=62970/100000 progress=63.0% loss=0.046722 lr=3.303199e-06 elapsed=23:20:55 avg_step=1.33s recent_step=1.38s eta=13:43:49 +2026-08-26 03:16:37 | INFO | train rank=0 | step=62980/100000 progress=63.0% loss=0.052917 lr=3.301644e-06 elapsed=23:21:07 avg_step=1.33s recent_step=1.18s eta=13:43:35 +2026-08-26 03:16:51 | INFO | train rank=0 | step=62990/100000 progress=63.0% loss=0.052246 lr=3.300089e-06 elapsed=23:21:21 avg_step=1.33s recent_step=1.43s eta=13:43:22 +2026-08-26 03:17:03 | INFO | train rank=0 | step=63000/100000 progress=63.0% loss=0.045105 lr=3.298534e-06 elapsed=23:21:33 avg_step=1.33s recent_step=1.19s eta=13:43:08 +2026-08-26 03:17:15 | INFO | train rank=0 | step=63010/100000 progress=63.0% loss=0.045166 lr=3.296979e-06 elapsed=23:21:45 avg_step=1.33s recent_step=1.25s eta=13:42:54 +2026-08-26 03:17:28 | INFO | train rank=0 | step=63020/100000 progress=63.0% loss=0.034538 lr=3.295425e-06 elapsed=23:21:58 avg_step=1.33s recent_step=1.27s eta=13:42:40 +2026-08-26 03:17:43 | INFO | train rank=0 | step=63030/100000 progress=63.0% loss=0.042450 lr=3.293871e-06 elapsed=23:22:13 avg_step=1.33s recent_step=1.55s eta=13:42:28 +2026-08-26 03:17:57 | INFO | train rank=0 | step=63040/100000 progress=63.0% loss=0.026680 lr=3.292316e-06 elapsed=23:22:27 avg_step=1.33s recent_step=1.39s eta=13:42:15 +2026-08-26 03:18:12 | INFO | train rank=0 | step=63050/100000 progress=63.0% loss=0.073242 lr=3.290763e-06 elapsed=23:22:42 avg_step=1.33s recent_step=1.46s eta=13:42:02 +2026-08-26 03:18:23 | INFO | train rank=0 | step=63060/100000 progress=63.1% loss=0.033661 lr=3.289209e-06 elapsed=23:22:53 avg_step=1.33s recent_step=1.11s eta=13:41:48 +2026-08-26 03:18:40 | INFO | train rank=0 | step=63070/100000 progress=63.1% loss=0.045090 lr=3.287655e-06 elapsed=23:23:10 avg_step=1.33s recent_step=1.66s eta=13:41:36 +2026-08-26 03:18:51 | INFO | train rank=0 | step=63080/100000 progress=63.1% loss=0.059692 lr=3.286102e-06 elapsed=23:23:21 avg_step=1.33s recent_step=1.16s eta=13:41:22 +2026-08-26 03:19:08 | INFO | train rank=0 | step=63090/100000 progress=63.1% loss=0.044991 lr=3.284549e-06 elapsed=23:23:38 avg_step=1.33s recent_step=1.67s eta=13:41:10 +2026-08-26 03:19:22 | INFO | train rank=0 | step=63100/100000 progress=63.1% loss=0.033203 lr=3.282996e-06 elapsed=23:23:51 avg_step=1.33s recent_step=1.35s eta=13:40:57 +2026-08-26 03:19:33 | INFO | train rank=0 | step=63110/100000 progress=63.1% loss=0.038361 lr=3.281443e-06 elapsed=23:24:03 avg_step=1.33s recent_step=1.11s eta=13:40:43 +2026-08-26 03:19:51 | INFO | train rank=0 | step=63120/100000 progress=63.1% loss=0.053802 lr=3.279890e-06 elapsed=23:24:21 avg_step=1.33s recent_step=1.82s eta=13:40:32 +2026-08-26 03:20:03 | INFO | train rank=0 | step=63130/100000 progress=63.1% loss=0.036240 lr=3.278338e-06 elapsed=23:24:32 avg_step=1.33s recent_step=1.17s eta=13:40:18 +2026-08-26 03:20:14 | INFO | train rank=0 | step=63140/100000 progress=63.1% loss=0.047134 lr=3.276785e-06 elapsed=23:24:44 avg_step=1.33s recent_step=1.14s eta=13:40:03 +2026-08-26 03:20:26 | INFO | train rank=0 | step=63150/100000 progress=63.1% loss=0.076294 lr=3.275233e-06 elapsed=23:24:56 avg_step=1.33s recent_step=1.22s eta=13:39:49 +2026-08-26 03:20:39 | INFO | train rank=0 | step=63160/100000 progress=63.2% loss=0.045761 lr=3.273681e-06 elapsed=23:25:09 avg_step=1.33s recent_step=1.30s eta=13:39:36 +2026-08-26 03:20:52 | INFO | train rank=0 | step=63170/100000 progress=63.2% loss=0.062500 lr=3.272130e-06 elapsed=23:25:22 avg_step=1.33s recent_step=1.32s eta=13:39:22 +2026-08-26 03:21:07 | INFO | train rank=0 | step=63180/100000 progress=63.2% loss=0.054749 lr=3.270578e-06 elapsed=23:25:37 avg_step=1.33s recent_step=1.48s eta=13:39:10 +2026-08-26 03:21:19 | INFO | train rank=0 | step=63190/100000 progress=63.2% loss=0.036003 lr=3.269027e-06 elapsed=23:25:49 avg_step=1.33s recent_step=1.15s eta=13:38:55 +2026-08-26 03:21:36 | INFO | train rank=0 | step=63200/100000 progress=63.2% loss=0.040848 lr=3.267476e-06 elapsed=23:26:06 avg_step=1.33s recent_step=1.71s eta=13:38:44 +2026-08-26 03:21:49 | INFO | train rank=0 | step=63210/100000 progress=63.2% loss=0.064636 lr=3.265925e-06 elapsed=23:26:19 avg_step=1.33s recent_step=1.30s eta=13:38:31 +2026-08-26 03:22:00 | INFO | train rank=0 | step=63220/100000 progress=63.2% loss=0.043900 lr=3.264374e-06 elapsed=23:26:30 avg_step=1.33s recent_step=1.15s eta=13:38:16 +2026-08-26 03:22:12 | INFO | train rank=0 | step=63230/100000 progress=63.2% loss=0.046875 lr=3.262824e-06 elapsed=23:26:42 avg_step=1.33s recent_step=1.17s eta=13:38:02 +2026-08-26 03:22:28 | INFO | train rank=0 | step=63240/100000 progress=63.2% loss=0.029884 lr=3.261273e-06 elapsed=23:26:58 avg_step=1.33s recent_step=1.58s eta=13:37:50 +2026-08-26 03:22:39 | INFO | train rank=0 | step=63250/100000 progress=63.2% loss=0.059219 lr=3.259723e-06 elapsed=23:27:09 avg_step=1.33s recent_step=1.15s eta=13:37:35 +2026-08-26 03:22:54 | INFO | train rank=0 | step=63260/100000 progress=63.3% loss=0.049347 lr=3.258173e-06 elapsed=23:27:24 avg_step=1.33s recent_step=1.51s eta=13:37:23 +2026-08-26 03:23:06 | INFO | train rank=0 | step=63270/100000 progress=63.3% loss=0.054619 lr=3.256623e-06 elapsed=23:27:36 avg_step=1.33s recent_step=1.19s eta=13:37:09 +2026-08-26 03:23:21 | INFO | train rank=0 | step=63280/100000 progress=63.3% loss=0.032364 lr=3.255074e-06 elapsed=23:27:50 avg_step=1.33s recent_step=1.42s eta=13:36:56 +2026-08-26 03:23:36 | INFO | train rank=0 | step=63290/100000 progress=63.3% loss=0.081604 lr=3.253524e-06 elapsed=23:28:06 avg_step=1.33s recent_step=1.56s eta=13:36:44 +2026-08-26 03:23:50 | INFO | train rank=0 | step=63300/100000 progress=63.3% loss=0.062592 lr=3.251975e-06 elapsed=23:28:20 avg_step=1.33s recent_step=1.39s eta=13:36:31 +2026-08-26 03:24:02 | INFO | train rank=0 | step=63310/100000 progress=63.3% loss=0.062042 lr=3.250426e-06 elapsed=23:28:32 avg_step=1.33s recent_step=1.18s eta=13:36:17 +2026-08-26 03:24:16 | INFO | train rank=0 | step=63320/100000 progress=63.3% loss=0.036163 lr=3.248877e-06 elapsed=23:28:46 avg_step=1.33s recent_step=1.42s eta=13:36:04 +2026-08-26 03:24:30 | INFO | train rank=0 | step=63330/100000 progress=63.3% loss=0.048492 lr=3.247328e-06 elapsed=23:29:00 avg_step=1.33s recent_step=1.43s eta=13:35:51 +2026-08-26 03:24:42 | INFO | train rank=0 | step=63340/100000 progress=63.3% loss=0.065277 lr=3.245780e-06 elapsed=23:29:12 avg_step=1.33s recent_step=1.20s eta=13:35:37 +2026-08-26 03:24:55 | INFO | train rank=0 | step=63350/100000 progress=63.3% loss=0.049164 lr=3.244232e-06 elapsed=23:29:24 avg_step=1.33s recent_step=1.22s eta=13:35:23 +2026-08-26 03:25:13 | INFO | train rank=0 | step=63360/100000 progress=63.4% loss=0.049652 lr=3.242684e-06 elapsed=23:29:43 avg_step=1.33s recent_step=1.84s eta=13:35:13 +2026-08-26 03:25:25 | INFO | train rank=0 | step=63370/100000 progress=63.4% loss=0.044724 lr=3.241136e-06 elapsed=23:29:54 avg_step=1.33s recent_step=1.17s eta=13:34:58 +2026-08-26 03:25:38 | INFO | train rank=0 | step=63380/100000 progress=63.4% loss=0.045013 lr=3.239588e-06 elapsed=23:30:08 avg_step=1.33s recent_step=1.36s eta=13:34:45 +2026-08-26 03:25:50 | INFO | train rank=0 | step=63390/100000 progress=63.4% loss=0.077362 lr=3.238041e-06 elapsed=23:30:20 avg_step=1.33s recent_step=1.16s eta=13:34:31 +2026-08-26 03:26:05 | INFO | train rank=0 | step=63400/100000 progress=63.4% loss=0.070312 lr=3.236493e-06 elapsed=23:30:35 avg_step=1.33s recent_step=1.53s eta=13:34:18 +2026-08-26 03:26:17 | INFO | train rank=0 | step=63410/100000 progress=63.4% loss=0.066162 lr=3.234946e-06 elapsed=23:30:46 avg_step=1.33s recent_step=1.15s eta=13:34:04 +2026-08-26 03:26:31 | INFO | train rank=0 | step=63420/100000 progress=63.4% loss=0.067139 lr=3.233399e-06 elapsed=23:31:01 avg_step=1.33s recent_step=1.42s eta=13:33:51 +2026-08-26 03:26:43 | INFO | train rank=0 | step=63430/100000 progress=63.4% loss=0.038452 lr=3.231852e-06 elapsed=23:31:13 avg_step=1.33s recent_step=1.21s eta=13:33:37 +2026-08-26 03:26:57 | INFO | train rank=0 | step=63440/100000 progress=63.4% loss=0.033875 lr=3.230306e-06 elapsed=23:31:27 avg_step=1.33s recent_step=1.47s eta=13:33:24 +2026-08-26 03:27:10 | INFO | train rank=0 | step=63450/100000 progress=63.4% loss=0.071991 lr=3.228760e-06 elapsed=23:31:39 avg_step=1.33s recent_step=1.21s eta=13:33:10 +2026-08-26 03:27:25 | INFO | train rank=0 | step=63460/100000 progress=63.5% loss=0.029762 lr=3.227213e-06 elapsed=23:31:55 avg_step=1.33s recent_step=1.57s eta=13:32:58 +2026-08-26 03:27:37 | INFO | train rank=0 | step=63470/100000 progress=63.5% loss=0.039734 lr=3.225667e-06 elapsed=23:32:07 avg_step=1.33s recent_step=1.15s eta=13:32:44 +2026-08-26 03:27:54 | INFO | train rank=0 | step=63480/100000 progress=63.5% loss=0.057648 lr=3.224122e-06 elapsed=23:32:24 avg_step=1.33s recent_step=1.76s eta=13:32:33 +2026-08-26 03:28:06 | INFO | train rank=0 | step=63490/100000 progress=63.5% loss=0.074188 lr=3.222576e-06 elapsed=23:32:36 avg_step=1.33s recent_step=1.18s eta=13:32:19 +2026-08-26 03:28:17 | INFO | train rank=0 | step=63500/100000 progress=63.5% loss=0.061203 lr=3.221031e-06 elapsed=23:32:47 avg_step=1.33s recent_step=1.12s eta=13:32:04 +2026-08-26 03:28:29 | INFO | train rank=0 | step=63510/100000 progress=63.5% loss=0.052475 lr=3.219486e-06 elapsed=23:32:59 avg_step=1.33s recent_step=1.20s eta=13:31:50 +2026-08-26 03:28:43 | INFO | train rank=0 | step=63520/100000 progress=63.5% loss=0.034714 lr=3.217941e-06 elapsed=23:33:13 avg_step=1.33s recent_step=1.37s eta=13:31:37 +2026-08-26 03:28:55 | INFO | train rank=0 | step=63530/100000 progress=63.5% loss=0.059830 lr=3.216396e-06 elapsed=23:33:25 avg_step=1.33s recent_step=1.18s eta=13:31:23 +2026-08-26 03:29:09 | INFO | train rank=0 | step=63540/100000 progress=63.5% loss=0.046310 lr=3.214851e-06 elapsed=23:33:39 avg_step=1.33s recent_step=1.41s eta=13:31:10 +2026-08-26 03:29:21 | INFO | train rank=0 | step=63550/100000 progress=63.5% loss=0.052841 lr=3.213307e-06 elapsed=23:33:51 avg_step=1.33s recent_step=1.18s eta=13:30:56 +2026-08-26 03:29:35 | INFO | train rank=0 | step=63560/100000 progress=63.6% loss=0.037567 lr=3.211763e-06 elapsed=23:34:05 avg_step=1.33s recent_step=1.42s eta=13:30:43 +2026-08-26 03:29:48 | INFO | train rank=0 | step=63570/100000 progress=63.6% loss=0.055786 lr=3.210219e-06 elapsed=23:34:18 avg_step=1.33s recent_step=1.30s eta=13:30:29 +2026-08-26 03:30:01 | INFO | train rank=0 | step=63580/100000 progress=63.6% loss=0.033188 lr=3.208675e-06 elapsed=23:34:31 avg_step=1.33s recent_step=1.35s eta=13:30:16 +2026-08-26 03:30:13 | INFO | train rank=0 | step=63590/100000 progress=63.6% loss=0.036095 lr=3.207131e-06 elapsed=23:34:43 avg_step=1.33s recent_step=1.14s eta=13:30:01 +2026-08-26 03:30:27 | INFO | train rank=0 | step=63600/100000 progress=63.6% loss=0.036057 lr=3.205588e-06 elapsed=23:34:57 avg_step=1.33s recent_step=1.40s eta=13:29:48 +2026-08-26 03:30:39 | INFO | train rank=0 | step=63610/100000 progress=63.6% loss=0.040894 lr=3.204045e-06 elapsed=23:35:09 avg_step=1.33s recent_step=1.19s eta=13:29:34 +2026-08-26 03:30:52 | INFO | train rank=0 | step=63620/100000 progress=63.6% loss=0.033806 lr=3.202502e-06 elapsed=23:35:22 avg_step=1.33s recent_step=1.37s eta=13:29:21 +2026-08-26 03:31:11 | INFO | train rank=0 | step=63630/100000 progress=63.6% loss=0.027878 lr=3.200959e-06 elapsed=23:35:41 avg_step=1.33s recent_step=1.84s eta=13:29:11 +2026-08-26 03:31:23 | INFO | train rank=0 | step=63640/100000 progress=63.6% loss=0.028961 lr=3.199416e-06 elapsed=23:35:53 avg_step=1.33s recent_step=1.20s eta=13:28:56 +2026-08-26 03:31:36 | INFO | train rank=0 | step=63650/100000 progress=63.6% loss=0.050598 lr=3.197874e-06 elapsed=23:36:06 avg_step=1.33s recent_step=1.34s eta=13:28:43 +2026-08-26 03:31:48 | INFO | train rank=0 | step=63660/100000 progress=63.7% loss=0.027374 lr=3.196331e-06 elapsed=23:36:18 avg_step=1.33s recent_step=1.19s eta=13:28:29 +2026-08-26 03:32:07 | INFO | train rank=0 | step=63670/100000 progress=63.7% loss=0.033432 lr=3.194789e-06 elapsed=23:36:37 avg_step=1.33s recent_step=1.92s eta=13:28:19 +2026-08-26 03:32:22 | INFO | train rank=0 | step=63680/100000 progress=63.7% loss=0.034485 lr=3.193248e-06 elapsed=23:36:52 avg_step=1.33s recent_step=1.51s eta=13:28:07 +2026-08-26 03:32:34 | INFO | train rank=0 | step=63690/100000 progress=63.7% loss=0.043800 lr=3.191706e-06 elapsed=23:37:03 avg_step=1.33s recent_step=1.14s eta=13:27:52 +2026-08-26 03:32:48 | INFO | train rank=0 | step=63700/100000 progress=63.7% loss=0.048752 lr=3.190164e-06 elapsed=23:37:18 avg_step=1.33s recent_step=1.49s eta=13:27:40 +2026-08-26 03:33:07 | INFO | train rank=0 | step=63710/100000 progress=63.7% loss=0.042175 lr=3.188623e-06 elapsed=23:37:37 avg_step=1.34s recent_step=1.83s eta=13:27:29 +2026-08-26 03:33:19 | INFO | train rank=0 | step=63720/100000 progress=63.7% loss=0.069885 lr=3.187082e-06 elapsed=23:37:49 avg_step=1.34s recent_step=1.26s eta=13:27:15 +2026-08-26 03:33:31 | INFO | train rank=0 | step=63730/100000 progress=63.7% loss=0.082520 lr=3.185541e-06 elapsed=23:38:01 avg_step=1.34s recent_step=1.19s eta=13:27:01 +2026-08-26 03:33:43 | INFO | train rank=0 | step=63740/100000 progress=63.7% loss=0.037216 lr=3.184001e-06 elapsed=23:38:13 avg_step=1.34s recent_step=1.17s eta=13:26:47 +2026-08-26 03:33:54 | INFO | train rank=0 | step=63750/100000 progress=63.7% loss=0.052032 lr=3.182460e-06 elapsed=23:38:24 avg_step=1.33s recent_step=1.13s eta=13:26:32 +2026-08-26 03:34:08 | INFO | train rank=0 | step=63760/100000 progress=63.8% loss=0.052650 lr=3.180920e-06 elapsed=23:38:38 avg_step=1.33s recent_step=1.39s eta=13:26:19 +2026-08-26 03:34:21 | INFO | train rank=0 | step=63770/100000 progress=63.8% loss=0.055176 lr=3.179380e-06 elapsed=23:38:51 avg_step=1.33s recent_step=1.30s eta=13:26:06 +2026-08-26 03:34:35 | INFO | train rank=0 | step=63780/100000 progress=63.8% loss=0.072083 lr=3.177840e-06 elapsed=23:39:05 avg_step=1.33s recent_step=1.37s eta=13:25:53 +2026-08-26 03:34:47 | INFO | train rank=0 | step=63790/100000 progress=63.8% loss=0.088715 lr=3.176300e-06 elapsed=23:39:17 avg_step=1.33s recent_step=1.23s eta=13:25:39 +2026-08-26 03:35:01 | INFO | train rank=0 | step=63800/100000 progress=63.8% loss=0.040085 lr=3.174761e-06 elapsed=23:39:31 avg_step=1.33s recent_step=1.40s eta=13:25:26 +2026-08-26 03:35:13 | INFO | train rank=0 | step=63810/100000 progress=63.8% loss=0.047943 lr=3.173222e-06 elapsed=23:39:43 avg_step=1.33s recent_step=1.17s eta=13:25:11 +2026-08-26 03:35:25 | INFO | train rank=0 | step=63820/100000 progress=63.8% loss=0.039185 lr=3.171683e-06 elapsed=23:39:55 avg_step=1.33s recent_step=1.18s eta=13:24:57 +2026-08-26 03:35:38 | INFO | train rank=0 | step=63830/100000 progress=63.8% loss=0.049194 lr=3.170144e-06 elapsed=23:40:08 avg_step=1.33s recent_step=1.33s eta=13:24:44 +2026-08-26 03:35:51 | INFO | train rank=0 | step=63840/100000 progress=63.8% loss=0.045944 lr=3.168605e-06 elapsed=23:40:21 avg_step=1.33s recent_step=1.28s eta=13:24:30 +2026-08-26 03:36:07 | INFO | train rank=0 | step=63850/100000 progress=63.8% loss=0.037140 lr=3.167067e-06 elapsed=23:40:37 avg_step=1.33s recent_step=1.60s eta=13:24:18 +2026-08-26 03:36:20 | INFO | train rank=0 | step=63860/100000 progress=63.9% loss=0.048080 lr=3.165528e-06 elapsed=23:40:50 avg_step=1.33s recent_step=1.30s eta=13:24:05 +2026-08-26 03:36:32 | INFO | train rank=0 | step=63870/100000 progress=63.9% loss=0.106262 lr=3.163990e-06 elapsed=23:41:02 avg_step=1.33s recent_step=1.20s eta=13:23:51 +2026-08-26 03:36:50 | INFO | train rank=0 | step=63880/100000 progress=63.9% loss=0.042801 lr=3.162452e-06 elapsed=23:41:20 avg_step=1.34s recent_step=1.82s eta=13:23:40 +2026-08-26 03:37:01 | INFO | train rank=0 | step=63890/100000 progress=63.9% loss=0.031319 lr=3.160915e-06 elapsed=23:41:31 avg_step=1.33s recent_step=1.14s eta=13:23:26 +2026-08-26 03:37:16 | INFO | train rank=0 | step=63900/100000 progress=63.9% loss=0.066925 lr=3.159377e-06 elapsed=23:41:46 avg_step=1.33s recent_step=1.44s eta=13:23:13 +2026-08-26 03:37:29 | INFO | train rank=0 | step=63910/100000 progress=63.9% loss=0.037781 lr=3.157840e-06 elapsed=23:41:59 avg_step=1.33s recent_step=1.31s eta=13:22:59 +2026-08-26 03:37:41 | INFO | train rank=0 | step=63920/100000 progress=63.9% loss=0.039703 lr=3.156303e-06 elapsed=23:42:11 avg_step=1.33s recent_step=1.18s eta=13:22:45 +2026-08-26 03:37:53 | INFO | train rank=0 | step=63930/100000 progress=63.9% loss=0.077087 lr=3.154766e-06 elapsed=23:42:23 avg_step=1.33s recent_step=1.22s eta=13:22:31 +2026-08-26 03:38:06 | INFO | train rank=0 | step=63940/100000 progress=63.9% loss=0.043877 lr=3.153229e-06 elapsed=23:42:36 avg_step=1.33s recent_step=1.29s eta=13:22:18 +2026-08-26 03:38:23 | INFO | train rank=0 | step=63950/100000 progress=63.9% loss=0.028572 lr=3.151693e-06 elapsed=23:42:53 avg_step=1.34s recent_step=1.75s eta=13:22:07 +2026-08-26 03:38:39 | INFO | train rank=0 | step=63960/100000 progress=64.0% loss=0.044930 lr=3.150157e-06 elapsed=23:43:08 avg_step=1.34s recent_step=1.51s eta=13:21:54 +2026-08-26 03:38:50 | INFO | train rank=0 | step=63970/100000 progress=64.0% loss=0.021812 lr=3.148621e-06 elapsed=23:43:20 avg_step=1.34s recent_step=1.16s eta=13:21:40 +2026-08-26 03:39:01 | INFO | train rank=0 | step=63980/100000 progress=64.0% loss=0.038864 lr=3.147085e-06 elapsed=23:43:31 avg_step=1.33s recent_step=1.13s eta=13:21:25 +2026-08-26 03:39:19 | INFO | train rank=0 | step=63990/100000 progress=64.0% loss=0.057251 lr=3.145549e-06 elapsed=23:43:49 avg_step=1.34s recent_step=1.81s eta=13:21:15 +2026-08-26 03:39:37 | INFO | train rank=0 | step=64000/100000 progress=64.0% loss=0.042404 lr=3.144014e-06 elapsed=23:44:07 avg_step=1.34s recent_step=1.78s eta=13:21:04 +2026-08-26 03:39:49 | INFO | train rank=0 | step=64010/100000 progress=64.0% loss=0.070953 lr=3.142479e-06 elapsed=23:44:19 avg_step=1.34s recent_step=1.16s eta=13:20:49 +2026-08-26 03:40:00 | INFO | train rank=0 | step=64020/100000 progress=64.0% loss=0.028778 lr=3.140943e-06 elapsed=23:44:30 avg_step=1.34s recent_step=1.16s eta=13:20:35 +2026-08-26 03:40:13 | INFO | train rank=0 | step=64030/100000 progress=64.0% loss=0.061859 lr=3.139409e-06 elapsed=23:44:43 avg_step=1.34s recent_step=1.26s eta=13:20:21 +2026-08-26 03:40:26 | INFO | train rank=0 | step=64040/100000 progress=64.0% loss=0.060211 lr=3.137874e-06 elapsed=23:44:56 avg_step=1.34s recent_step=1.31s eta=13:20:08 +2026-08-26 03:40:39 | INFO | train rank=0 | step=64050/100000 progress=64.0% loss=0.052216 lr=3.136340e-06 elapsed=23:45:09 avg_step=1.34s recent_step=1.30s eta=13:19:54 +2026-08-26 03:40:57 | INFO | train rank=0 | step=64060/100000 progress=64.1% loss=0.047333 lr=3.134805e-06 elapsed=23:45:27 avg_step=1.34s recent_step=1.82s eta=13:19:44 +2026-08-26 03:41:09 | INFO | train rank=0 | step=64070/100000 progress=64.1% loss=0.039253 lr=3.133271e-06 elapsed=23:45:39 avg_step=1.34s recent_step=1.16s eta=13:19:29 +2026-08-26 03:41:22 | INFO | train rank=0 | step=64080/100000 progress=64.1% loss=0.025837 lr=3.131738e-06 elapsed=23:45:52 avg_step=1.34s recent_step=1.29s eta=13:19:16 +2026-08-26 03:41:37 | INFO | train rank=0 | step=64090/100000 progress=64.1% loss=0.058960 lr=3.130204e-06 elapsed=23:46:07 avg_step=1.34s recent_step=1.47s eta=13:19:03 +2026-08-26 03:41:48 | INFO | train rank=0 | step=64100/100000 progress=64.1% loss=0.044128 lr=3.128671e-06 elapsed=23:46:18 avg_step=1.34s recent_step=1.12s eta=13:18:49 +2026-08-26 03:42:04 | INFO | train rank=0 | step=64110/100000 progress=64.1% loss=0.039963 lr=3.127137e-06 elapsed=23:46:34 avg_step=1.34s recent_step=1.61s eta=13:18:37 +2026-08-26 03:42:18 | INFO | train rank=0 | step=64120/100000 progress=64.1% loss=0.063049 lr=3.125604e-06 elapsed=23:46:48 avg_step=1.34s recent_step=1.42s eta=13:18:24 +2026-08-26 03:42:29 | INFO | train rank=0 | step=64130/100000 progress=64.1% loss=0.044861 lr=3.124072e-06 elapsed=23:46:59 avg_step=1.34s recent_step=1.13s eta=13:18:09 +2026-08-26 03:42:44 | INFO | train rank=0 | step=64140/100000 progress=64.1% loss=0.040909 lr=3.122539e-06 elapsed=23:47:14 avg_step=1.34s recent_step=1.44s eta=13:17:57 +2026-08-26 03:42:59 | INFO | train rank=0 | step=64150/100000 progress=64.1% loss=0.039047 lr=3.121007e-06 elapsed=23:47:29 avg_step=1.34s recent_step=1.57s eta=13:17:45 +2026-08-26 03:43:14 | INFO | train rank=0 | step=64160/100000 progress=64.2% loss=0.048935 lr=3.119474e-06 elapsed=23:47:44 avg_step=1.34s recent_step=1.50s eta=13:17:32 +2026-08-26 03:43:26 | INFO | train rank=0 | step=64170/100000 progress=64.2% loss=0.033173 lr=3.117943e-06 elapsed=23:47:56 avg_step=1.34s recent_step=1.19s eta=13:17:18 +2026-08-26 03:43:37 | INFO | train rank=0 | step=64180/100000 progress=64.2% loss=0.046509 lr=3.116411e-06 elapsed=23:48:07 avg_step=1.34s recent_step=1.11s eta=13:17:03 +2026-08-26 03:43:52 | INFO | train rank=0 | step=64190/100000 progress=64.2% loss=0.064972 lr=3.114879e-06 elapsed=23:48:22 avg_step=1.34s recent_step=1.49s eta=13:16:51 +2026-08-26 03:44:07 | INFO | train rank=0 | step=64200/100000 progress=64.2% loss=0.072464 lr=3.113348e-06 elapsed=23:48:37 avg_step=1.34s recent_step=1.51s eta=13:16:39 +2026-08-26 03:44:18 | INFO | train rank=0 | step=64210/100000 progress=64.2% loss=0.056000 lr=3.111817e-06 elapsed=23:48:48 avg_step=1.34s recent_step=1.10s eta=13:16:24 +2026-08-26 03:44:32 | INFO | train rank=0 | step=64220/100000 progress=64.2% loss=0.046249 lr=3.110286e-06 elapsed=23:49:02 avg_step=1.34s recent_step=1.36s eta=13:16:11 +2026-08-26 03:44:44 | INFO | train rank=0 | step=64230/100000 progress=64.2% loss=0.044632 lr=3.108755e-06 elapsed=23:49:14 avg_step=1.34s recent_step=1.17s eta=13:15:56 +2026-08-26 03:44:56 | INFO | train rank=0 | step=64240/100000 progress=64.2% loss=0.063599 lr=3.107225e-06 elapsed=23:49:26 avg_step=1.34s recent_step=1.24s eta=13:15:42 +2026-08-26 03:45:10 | INFO | train rank=0 | step=64250/100000 progress=64.2% loss=0.036896 lr=3.105694e-06 elapsed=23:49:40 avg_step=1.34s recent_step=1.41s eta=13:15:30 +2026-08-26 03:45:29 | INFO | train rank=0 | step=64260/100000 progress=64.3% loss=0.040253 lr=3.104164e-06 elapsed=23:49:59 avg_step=1.34s recent_step=1.89s eta=13:15:19 +2026-08-26 03:45:41 | INFO | train rank=0 | step=64270/100000 progress=64.3% loss=0.046982 lr=3.102634e-06 elapsed=23:50:10 avg_step=1.34s recent_step=1.15s eta=13:15:05 +2026-08-26 03:45:53 | INFO | train rank=0 | step=64280/100000 progress=64.3% loss=0.061035 lr=3.101105e-06 elapsed=23:50:23 avg_step=1.34s recent_step=1.28s eta=13:14:51 +2026-08-26 03:46:05 | INFO | train rank=0 | step=64290/100000 progress=64.3% loss=0.057404 lr=3.099575e-06 elapsed=23:50:35 avg_step=1.34s recent_step=1.13s eta=13:14:37 +2026-08-26 03:46:18 | INFO | train rank=0 | step=64300/100000 progress=64.3% loss=0.035568 lr=3.098046e-06 elapsed=23:50:48 avg_step=1.34s recent_step=1.35s eta=13:14:24 +2026-08-26 03:46:35 | INFO | train rank=0 | step=64310/100000 progress=64.3% loss=0.054337 lr=3.096517e-06 elapsed=23:51:05 avg_step=1.34s recent_step=1.66s eta=13:14:12 +2026-08-26 03:46:47 | INFO | train rank=0 | step=64320/100000 progress=64.3% loss=0.061798 lr=3.094988e-06 elapsed=23:51:17 avg_step=1.34s recent_step=1.18s eta=13:13:58 +2026-08-26 03:47:01 | INFO | train rank=0 | step=64330/100000 progress=64.3% loss=0.082611 lr=3.093459e-06 elapsed=23:51:31 avg_step=1.34s recent_step=1.41s eta=13:13:45 +2026-08-26 03:47:12 | INFO | train rank=0 | step=64340/100000 progress=64.3% loss=0.091232 lr=3.091931e-06 elapsed=23:51:42 avg_step=1.34s recent_step=1.16s eta=13:13:30 +2026-08-26 03:47:29 | INFO | train rank=0 | step=64350/100000 progress=64.3% loss=0.048706 lr=3.090403e-06 elapsed=23:51:59 avg_step=1.34s recent_step=1.64s eta=13:13:19 +2026-08-26 03:47:41 | INFO | train rank=0 | step=64360/100000 progress=64.4% loss=0.040482 lr=3.088874e-06 elapsed=23:52:10 avg_step=1.34s recent_step=1.19s eta=13:13:05 +2026-08-26 03:47:55 | INFO | train rank=0 | step=64370/100000 progress=64.4% loss=0.056519 lr=3.087347e-06 elapsed=23:52:24 avg_step=1.34s recent_step=1.40s eta=13:12:52 +2026-08-26 03:48:10 | INFO | train rank=0 | step=64380/100000 progress=64.4% loss=0.060181 lr=3.085819e-06 elapsed=23:52:40 avg_step=1.34s recent_step=1.57s eta=13:12:40 +2026-08-26 03:48:21 | INFO | train rank=0 | step=64390/100000 progress=64.4% loss=0.044647 lr=3.084292e-06 elapsed=23:52:51 avg_step=1.34s recent_step=1.12s eta=13:12:25 +2026-08-26 03:48:34 | INFO | train rank=0 | step=64400/100000 progress=64.4% loss=0.036499 lr=3.082764e-06 elapsed=23:53:04 avg_step=1.34s recent_step=1.24s eta=13:12:11 +2026-08-26 03:48:48 | INFO | train rank=0 | step=64410/100000 progress=64.4% loss=0.044502 lr=3.081237e-06 elapsed=23:53:18 avg_step=1.34s recent_step=1.41s eta=13:11:58 +2026-08-26 03:49:00 | INFO | train rank=0 | step=64420/100000 progress=64.4% loss=0.041000 lr=3.079711e-06 elapsed=23:53:30 avg_step=1.34s recent_step=1.17s eta=13:11:44 +2026-08-26 03:49:17 | INFO | train rank=0 | step=64430/100000 progress=64.4% loss=0.059738 lr=3.078184e-06 elapsed=23:53:47 avg_step=1.34s recent_step=1.75s eta=13:11:33 +2026-08-26 03:49:29 | INFO | train rank=0 | step=64440/100000 progress=64.4% loss=0.031143 lr=3.076658e-06 elapsed=23:53:59 avg_step=1.34s recent_step=1.15s eta=13:11:19 +2026-08-26 03:49:40 | INFO | train rank=0 | step=64450/100000 progress=64.5% loss=0.075912 lr=3.075132e-06 elapsed=23:54:10 avg_step=1.34s recent_step=1.16s eta=13:11:04 +2026-08-26 03:49:55 | INFO | train rank=0 | step=64460/100000 progress=64.5% loss=0.037567 lr=3.073606e-06 elapsed=23:54:25 avg_step=1.34s recent_step=1.44s eta=13:10:52 +2026-08-26 03:50:08 | INFO | train rank=0 | step=64470/100000 progress=64.5% loss=0.036026 lr=3.072080e-06 elapsed=23:54:37 avg_step=1.34s recent_step=1.28s eta=13:10:38 +2026-08-26 03:50:19 | INFO | train rank=0 | step=64480/100000 progress=64.5% loss=0.050354 lr=3.070555e-06 elapsed=23:54:49 avg_step=1.34s recent_step=1.17s eta=13:10:24 +2026-08-26 03:50:38 | INFO | train rank=0 | step=64490/100000 progress=64.5% loss=0.054062 lr=3.069029e-06 elapsed=23:55:08 avg_step=1.34s recent_step=1.86s eta=13:10:13 +2026-08-26 03:50:49 | INFO | train rank=0 | step=64500/100000 progress=64.5% loss=0.046188 lr=3.067504e-06 elapsed=23:55:19 avg_step=1.34s recent_step=1.15s eta=13:09:59 +2026-08-26 03:51:01 | INFO | train rank=0 | step=64510/100000 progress=64.5% loss=0.043762 lr=3.065979e-06 elapsed=23:55:31 avg_step=1.34s recent_step=1.19s eta=13:09:45 +2026-08-26 03:51:13 | INFO | train rank=0 | step=64520/100000 progress=64.5% loss=0.045273 lr=3.064455e-06 elapsed=23:55:43 avg_step=1.34s recent_step=1.17s eta=13:09:30 +2026-08-26 03:51:24 | INFO | train rank=0 | step=64530/100000 progress=64.5% loss=0.063507 lr=3.062930e-06 elapsed=23:55:54 avg_step=1.34s recent_step=1.16s eta=13:09:16 +2026-08-26 03:51:38 | INFO | train rank=0 | step=64540/100000 progress=64.5% loss=0.062592 lr=3.061406e-06 elapsed=23:56:08 avg_step=1.34s recent_step=1.38s eta=13:09:03 +2026-08-26 03:51:58 | INFO | train rank=0 | step=64550/100000 progress=64.5% loss=0.057419 lr=3.059882e-06 elapsed=23:56:28 avg_step=1.34s recent_step=1.94s eta=13:08:53 +2026-08-26 03:52:09 | INFO | train rank=0 | step=64560/100000 progress=64.6% loss=0.060486 lr=3.058358e-06 elapsed=23:56:39 avg_step=1.34s recent_step=1.16s eta=13:08:39 +2026-08-26 03:52:23 | INFO | train rank=0 | step=64570/100000 progress=64.6% loss=0.131836 lr=3.056834e-06 elapsed=23:56:52 avg_step=1.34s recent_step=1.33s eta=13:08:25 +2026-08-26 03:52:34 | INFO | train rank=0 | step=64580/100000 progress=64.6% loss=0.041626 lr=3.055311e-06 elapsed=23:57:04 avg_step=1.34s recent_step=1.15s eta=13:08:11 +2026-08-26 03:52:51 | INFO | train rank=0 | step=64590/100000 progress=64.6% loss=0.068634 lr=3.053788e-06 elapsed=23:57:21 avg_step=1.34s recent_step=1.69s eta=13:07:59 +2026-08-26 03:53:04 | INFO | train rank=0 | step=64600/100000 progress=64.6% loss=0.043396 lr=3.052265e-06 elapsed=23:57:34 avg_step=1.34s recent_step=1.31s eta=13:07:46 +2026-08-26 03:53:17 | INFO | train rank=0 | step=64610/100000 progress=64.6% loss=0.054764 lr=3.050742e-06 elapsed=23:57:47 avg_step=1.34s recent_step=1.31s eta=13:07:32 +2026-08-26 03:53:29 | INFO | train rank=0 | step=64620/100000 progress=64.6% loss=0.027451 lr=3.049220e-06 elapsed=23:57:59 avg_step=1.34s recent_step=1.16s eta=13:07:18 +2026-08-26 03:53:40 | INFO | train rank=0 | step=64630/100000 progress=64.6% loss=0.076202 lr=3.047697e-06 elapsed=23:58:10 avg_step=1.34s recent_step=1.13s eta=13:07:04 +2026-08-26 03:53:52 | INFO | train rank=0 | step=64640/100000 progress=64.6% loss=0.029266 lr=3.046175e-06 elapsed=23:58:22 avg_step=1.34s recent_step=1.21s eta=13:06:50 +2026-08-26 03:54:07 | INFO | train rank=0 | step=64650/100000 progress=64.6% loss=0.062775 lr=3.044653e-06 elapsed=23:58:37 avg_step=1.34s recent_step=1.54s eta=13:06:37 +2026-08-26 03:54:19 | INFO | train rank=0 | step=64660/100000 progress=64.7% loss=0.075592 lr=3.043132e-06 elapsed=23:58:49 avg_step=1.34s recent_step=1.20s eta=13:06:23 +2026-08-26 03:54:39 | INFO | train rank=0 | step=64670/100000 progress=64.7% loss=0.082397 lr=3.041610e-06 elapsed=23:59:08 avg_step=1.34s recent_step=1.91s eta=13:06:13 +2026-08-26 03:54:50 | INFO | train rank=0 | step=64680/100000 progress=64.7% loss=0.058838 lr=3.040089e-06 elapsed=23:59:20 avg_step=1.34s recent_step=1.16s eta=13:05:59 +2026-08-26 03:55:02 | INFO | train rank=0 | step=64690/100000 progress=64.7% loss=0.126526 lr=3.038568e-06 elapsed=23:59:32 avg_step=1.34s recent_step=1.23s eta=13:05:45 +2026-08-26 03:55:19 | INFO | train rank=0 | step=64700/100000 progress=64.7% loss=0.033981 lr=3.037047e-06 elapsed=23:59:49 avg_step=1.34s recent_step=1.69s eta=13:05:33 +2026-08-26 03:55:31 | INFO | train rank=0 | step=64710/100000 progress=64.7% loss=0.061615 lr=3.035526e-06 elapsed=24:00:01 avg_step=1.34s recent_step=1.20s eta=13:05:19 +2026-08-26 03:55:43 | INFO | train rank=0 | step=64720/100000 progress=64.7% loss=0.062012 lr=3.034006e-06 elapsed=24:00:13 avg_step=1.34s recent_step=1.22s eta=13:05:05 +2026-08-26 03:55:55 | INFO | train rank=0 | step=64730/100000 progress=64.7% loss=0.041061 lr=3.032486e-06 elapsed=24:00:25 avg_step=1.34s recent_step=1.16s eta=13:04:51 +2026-08-26 03:56:08 | INFO | train rank=0 | step=64740/100000 progress=64.7% loss=0.051086 lr=3.030966e-06 elapsed=24:00:38 avg_step=1.34s recent_step=1.28s eta=13:04:37 +2026-08-26 03:56:21 | INFO | train rank=0 | step=64750/100000 progress=64.8% loss=0.056602 lr=3.029446e-06 elapsed=24:00:51 avg_step=1.34s recent_step=1.28s eta=13:04:24 +2026-08-26 03:56:33 | INFO | train rank=0 | step=64760/100000 progress=64.8% loss=0.049103 lr=3.027927e-06 elapsed=24:01:02 avg_step=1.34s recent_step=1.18s eta=13:04:09 +2026-08-26 03:56:47 | INFO | train rank=0 | step=64770/100000 progress=64.8% loss=0.061249 lr=3.026407e-06 elapsed=24:01:17 avg_step=1.34s recent_step=1.44s eta=13:03:57 +2026-08-26 03:57:02 | INFO | train rank=0 | step=64780/100000 progress=64.8% loss=0.036575 lr=3.024888e-06 elapsed=24:01:32 avg_step=1.34s recent_step=1.54s eta=13:03:44 +2026-08-26 03:57:14 | INFO | train rank=0 | step=64790/100000 progress=64.8% loss=0.026367 lr=3.023369e-06 elapsed=24:01:44 avg_step=1.34s recent_step=1.20s eta=13:03:30 +2026-08-26 03:57:31 | INFO | train rank=0 | step=64800/100000 progress=64.8% loss=0.032204 lr=3.021851e-06 elapsed=24:02:01 avg_step=1.34s recent_step=1.71s eta=13:03:19 +2026-08-26 03:57:47 | INFO | train rank=0 | step=64810/100000 progress=64.8% loss=0.030014 lr=3.020332e-06 elapsed=24:02:17 avg_step=1.34s recent_step=1.59s eta=13:03:07 +2026-08-26 03:58:02 | INFO | train rank=0 | step=64820/100000 progress=64.8% loss=0.062897 lr=3.018814e-06 elapsed=24:02:32 avg_step=1.34s recent_step=1.46s eta=13:02:54 +2026-08-26 03:58:13 | INFO | train rank=0 | step=64830/100000 progress=64.8% loss=0.046814 lr=3.017296e-06 elapsed=24:02:43 avg_step=1.34s recent_step=1.13s eta=13:02:40 +2026-08-26 03:58:26 | INFO | train rank=0 | step=64840/100000 progress=64.8% loss=0.053513 lr=3.015778e-06 elapsed=24:02:56 avg_step=1.34s recent_step=1.33s eta=13:02:26 +2026-08-26 03:58:38 | INFO | train rank=0 | step=64850/100000 progress=64.8% loss=0.055069 lr=3.014261e-06 elapsed=24:03:08 avg_step=1.34s recent_step=1.20s eta=13:02:12 +2026-08-26 03:58:52 | INFO | train rank=0 | step=64860/100000 progress=64.9% loss=0.057846 lr=3.012743e-06 elapsed=24:03:22 avg_step=1.34s recent_step=1.40s eta=13:01:59 +2026-08-26 03:59:07 | INFO | train rank=0 | step=64870/100000 progress=64.9% loss=0.051620 lr=3.011226e-06 elapsed=24:03:37 avg_step=1.34s recent_step=1.49s eta=13:01:47 +2026-08-26 03:59:19 | INFO | train rank=0 | step=64880/100000 progress=64.9% loss=0.049026 lr=3.009709e-06 elapsed=24:03:49 avg_step=1.34s recent_step=1.13s eta=13:01:32 +2026-08-26 03:59:33 | INFO | train rank=0 | step=64890/100000 progress=64.9% loss=0.032391 lr=3.008192e-06 elapsed=24:04:03 avg_step=1.34s recent_step=1.41s eta=13:01:19 +2026-08-26 03:59:45 | INFO | train rank=0 | step=64900/100000 progress=64.9% loss=0.096130 lr=3.006676e-06 elapsed=24:04:15 avg_step=1.34s recent_step=1.22s eta=13:01:05 +2026-08-26 03:59:56 | INFO | train rank=0 | step=64910/100000 progress=64.9% loss=0.047340 lr=3.005160e-06 elapsed=24:04:26 avg_step=1.34s recent_step=1.11s eta=13:00:51 +2026-08-26 04:00:10 | INFO | train rank=0 | step=64920/100000 progress=64.9% loss=0.063843 lr=3.003644e-06 elapsed=24:04:40 avg_step=1.34s recent_step=1.38s eta=13:00:38 +2026-08-26 04:00:25 | INFO | train rank=0 | step=64930/100000 progress=64.9% loss=0.055275 lr=3.002128e-06 elapsed=24:04:55 avg_step=1.34s recent_step=1.52s eta=13:00:25 +2026-08-26 04:00:36 | INFO | train rank=0 | step=64940/100000 progress=64.9% loss=0.076874 lr=3.000612e-06 elapsed=24:05:06 avg_step=1.34s recent_step=1.15s eta=13:00:11 +2026-08-26 04:00:51 | INFO | train rank=0 | step=64950/100000 progress=65.0% loss=0.047897 lr=2.999097e-06 elapsed=24:05:21 avg_step=1.34s recent_step=1.44s eta=12:59:58 +2026-08-26 04:01:06 | INFO | train rank=0 | step=64960/100000 progress=65.0% loss=0.052216 lr=2.997581e-06 elapsed=24:05:36 avg_step=1.34s recent_step=1.49s eta=12:59:46 +2026-08-26 04:01:19 | INFO | train rank=0 | step=64970/100000 progress=65.0% loss=0.043274 lr=2.996066e-06 elapsed=24:05:49 avg_step=1.34s recent_step=1.37s eta=12:59:33 +2026-08-26 04:01:35 | INFO | train rank=0 | step=64980/100000 progress=65.0% loss=0.092316 lr=2.994552e-06 elapsed=24:06:05 avg_step=1.34s recent_step=1.59s eta=12:59:21 +2026-08-26 04:01:46 | INFO | train rank=0 | step=64990/100000 progress=65.0% loss=0.034882 lr=2.993037e-06 elapsed=24:06:16 avg_step=1.34s recent_step=1.11s eta=12:59:06 +2026-08-26 04:02:04 | INFO | train rank=0 | step=65000/100000 progress=65.0% loss=0.033424 lr=2.991523e-06 elapsed=24:06:34 avg_step=1.34s recent_step=1.75s eta=12:58:55 +2026-08-26 04:02:15 | INFO | train rank=0 | step=65010/100000 progress=65.0% loss=0.047745 lr=2.990009e-06 elapsed=24:06:45 avg_step=1.34s recent_step=1.12s eta=12:58:40 +2026-08-26 04:02:27 | INFO | train rank=0 | step=65020/100000 progress=65.0% loss=0.041153 lr=2.988495e-06 elapsed=24:06:57 avg_step=1.34s recent_step=1.16s eta=12:58:26 +2026-08-26 04:02:39 | INFO | train rank=0 | step=65030/100000 progress=65.0% loss=0.065094 lr=2.986981e-06 elapsed=24:07:09 avg_step=1.34s recent_step=1.19s eta=12:58:12 +2026-08-26 04:02:50 | INFO | train rank=0 | step=65040/100000 progress=65.0% loss=0.061676 lr=2.985468e-06 elapsed=24:07:20 avg_step=1.34s recent_step=1.10s eta=12:57:57 +2026-08-26 04:03:02 | INFO | train rank=0 | step=65050/100000 progress=65.0% loss=0.037949 lr=2.983955e-06 elapsed=24:07:32 avg_step=1.34s recent_step=1.23s eta=12:57:43 +2026-08-26 04:03:14 | INFO | train rank=0 | step=65060/100000 progress=65.1% loss=0.063843 lr=2.982442e-06 elapsed=24:07:44 avg_step=1.34s recent_step=1.18s eta=12:57:29 +2026-08-26 04:03:26 | INFO | train rank=0 | step=65070/100000 progress=65.1% loss=0.050995 lr=2.980929e-06 elapsed=24:07:56 avg_step=1.34s recent_step=1.21s eta=12:57:15 +2026-08-26 04:03:43 | INFO | train rank=0 | step=65080/100000 progress=65.1% loss=0.052505 lr=2.979416e-06 elapsed=24:08:12 avg_step=1.34s recent_step=1.66s eta=12:57:04 +2026-08-26 04:03:58 | INFO | train rank=0 | step=65090/100000 progress=65.1% loss=0.040359 lr=2.977904e-06 elapsed=24:08:28 avg_step=1.34s recent_step=1.56s eta=12:56:52 +2026-08-26 04:04:10 | INFO | train rank=0 | step=65100/100000 progress=65.1% loss=0.055145 lr=2.976392e-06 elapsed=24:08:39 avg_step=1.34s recent_step=1.14s eta=12:56:37 +2026-08-26 04:04:25 | INFO | train rank=0 | step=65110/100000 progress=65.1% loss=0.037369 lr=2.974880e-06 elapsed=24:08:55 avg_step=1.34s recent_step=1.53s eta=12:56:25 +2026-08-26 04:04:37 | INFO | train rank=0 | step=65120/100000 progress=65.1% loss=0.035524 lr=2.973368e-06 elapsed=24:09:07 avg_step=1.34s recent_step=1.25s eta=12:56:11 +2026-08-26 04:04:49 | INFO | train rank=0 | step=65130/100000 progress=65.1% loss=0.056610 lr=2.971857e-06 elapsed=24:09:19 avg_step=1.34s recent_step=1.17s eta=12:55:57 +2026-08-26 04:05:03 | INFO | train rank=0 | step=65140/100000 progress=65.1% loss=0.060944 lr=2.970346e-06 elapsed=24:09:33 avg_step=1.34s recent_step=1.36s eta=12:55:44 +2026-08-26 04:05:17 | INFO | train rank=0 | step=65150/100000 progress=65.1% loss=0.038864 lr=2.968835e-06 elapsed=24:09:46 avg_step=1.34s recent_step=1.39s eta=12:55:31 +2026-08-26 04:05:30 | INFO | train rank=0 | step=65160/100000 progress=65.2% loss=0.042145 lr=2.967324e-06 elapsed=24:09:59 avg_step=1.34s recent_step=1.30s eta=12:55:17 +2026-08-26 04:05:45 | INFO | train rank=0 | step=65170/100000 progress=65.2% loss=0.065353 lr=2.965813e-06 elapsed=24:10:15 avg_step=1.34s recent_step=1.55s eta=12:55:05 +2026-08-26 04:06:00 | INFO | train rank=0 | step=65180/100000 progress=65.2% loss=0.067902 lr=2.964303e-06 elapsed=24:10:30 avg_step=1.34s recent_step=1.50s eta=12:54:52 +2026-08-26 04:06:12 | INFO | train rank=0 | step=65190/100000 progress=65.2% loss=0.038132 lr=2.962793e-06 elapsed=24:10:42 avg_step=1.34s recent_step=1.23s eta=12:54:38 +2026-08-26 04:06:26 | INFO | train rank=0 | step=65200/100000 progress=65.2% loss=0.036285 lr=2.961283e-06 elapsed=24:10:56 avg_step=1.34s recent_step=1.37s eta=12:54:25 +2026-08-26 04:06:38 | INFO | train rank=0 | step=65210/100000 progress=65.2% loss=0.037613 lr=2.959773e-06 elapsed=24:11:07 avg_step=1.34s recent_step=1.16s eta=12:54:11 +2026-08-26 04:06:53 | INFO | train rank=0 | step=65220/100000 progress=65.2% loss=0.036980 lr=2.958264e-06 elapsed=24:11:23 avg_step=1.34s recent_step=1.52s eta=12:53:58 +2026-08-26 04:07:09 | INFO | train rank=0 | step=65230/100000 progress=65.2% loss=0.036301 lr=2.956755e-06 elapsed=24:11:39 avg_step=1.34s recent_step=1.67s eta=12:53:47 +2026-08-26 04:07:24 | INFO | train rank=0 | step=65240/100000 progress=65.2% loss=0.051941 lr=2.955246e-06 elapsed=24:11:54 avg_step=1.34s recent_step=1.48s eta=12:53:34 +2026-08-26 04:07:35 | INFO | train rank=0 | step=65250/100000 progress=65.2% loss=0.057388 lr=2.953737e-06 elapsed=24:12:05 avg_step=1.34s recent_step=1.12s eta=12:53:20 +2026-08-26 04:07:47 | INFO | train rank=0 | step=65260/100000 progress=65.3% loss=0.027603 lr=2.952228e-06 elapsed=24:12:17 avg_step=1.34s recent_step=1.18s eta=12:53:06 +2026-08-26 04:08:01 | INFO | train rank=0 | step=65270/100000 progress=65.3% loss=0.054764 lr=2.950720e-06 elapsed=24:12:31 avg_step=1.34s recent_step=1.37s eta=12:52:53 +2026-08-26 04:08:16 | INFO | train rank=0 | step=65280/100000 progress=65.3% loss=0.069290 lr=2.949212e-06 elapsed=24:12:46 avg_step=1.34s recent_step=1.54s eta=12:52:40 +2026-08-26 04:08:31 | INFO | train rank=0 | step=65290/100000 progress=65.3% loss=0.047699 lr=2.947704e-06 elapsed=24:13:01 avg_step=1.34s recent_step=1.47s eta=12:52:28 +2026-08-26 04:08:43 | INFO | train rank=0 | step=65300/100000 progress=65.3% loss=0.064682 lr=2.946196e-06 elapsed=24:13:13 avg_step=1.34s recent_step=1.17s eta=12:52:13 +2026-08-26 04:08:55 | INFO | train rank=0 | step=65310/100000 progress=65.3% loss=0.049133 lr=2.944689e-06 elapsed=24:13:25 avg_step=1.34s recent_step=1.20s eta=12:51:59 +2026-08-26 04:09:07 | INFO | train rank=0 | step=65320/100000 progress=65.3% loss=0.063049 lr=2.943182e-06 elapsed=24:13:36 avg_step=1.34s recent_step=1.18s eta=12:51:45 +2026-08-26 04:09:26 | INFO | train rank=0 | step=65330/100000 progress=65.3% loss=0.040710 lr=2.941675e-06 elapsed=24:13:56 avg_step=1.34s recent_step=1.93s eta=12:51:35 +2026-08-26 04:09:38 | INFO | train rank=0 | step=65340/100000 progress=65.3% loss=0.049980 lr=2.940168e-06 elapsed=24:14:08 avg_step=1.34s recent_step=1.19s eta=12:51:21 +2026-08-26 04:09:51 | INFO | train rank=0 | step=65350/100000 progress=65.3% loss=0.054276 lr=2.938661e-06 elapsed=24:14:20 avg_step=1.34s recent_step=1.28s eta=12:51:07 +2026-08-26 04:10:02 | INFO | train rank=0 | step=65360/100000 progress=65.4% loss=0.100769 lr=2.937155e-06 elapsed=24:14:32 avg_step=1.34s recent_step=1.17s eta=12:50:53 +2026-08-26 04:10:21 | INFO | train rank=0 | step=65370/100000 progress=65.4% loss=0.040482 lr=2.935649e-06 elapsed=24:14:51 avg_step=1.34s recent_step=1.86s eta=12:50:42 +2026-08-26 04:10:32 | INFO | train rank=0 | step=65380/100000 progress=65.4% loss=0.054077 lr=2.934143e-06 elapsed=24:15:02 avg_step=1.34s recent_step=1.12s eta=12:50:28 +2026-08-26 04:10:47 | INFO | train rank=0 | step=65390/100000 progress=65.4% loss=0.094788 lr=2.932638e-06 elapsed=24:15:16 avg_step=1.34s recent_step=1.45s eta=12:50:15 +2026-08-26 04:11:02 | INFO | train rank=0 | step=65400/100000 progress=65.4% loss=0.036591 lr=2.931132e-06 elapsed=24:15:32 avg_step=1.34s recent_step=1.53s eta=12:50:03 +2026-08-26 04:11:14 | INFO | train rank=0 | step=65410/100000 progress=65.4% loss=0.058151 lr=2.929627e-06 elapsed=24:15:44 avg_step=1.34s recent_step=1.25s eta=12:49:49 +2026-08-26 04:11:28 | INFO | train rank=0 | step=65420/100000 progress=65.4% loss=0.045609 lr=2.928122e-06 elapsed=24:15:58 avg_step=1.34s recent_step=1.41s eta=12:49:36 +2026-08-26 04:11:40 | INFO | train rank=0 | step=65430/100000 progress=65.4% loss=0.104797 lr=2.926617e-06 elapsed=24:16:10 avg_step=1.34s recent_step=1.15s eta=12:49:22 +2026-08-26 04:11:53 | INFO | train rank=0 | step=65440/100000 progress=65.4% loss=0.051239 lr=2.925113e-06 elapsed=24:16:23 avg_step=1.34s recent_step=1.30s eta=12:49:08 +2026-08-26 04:12:08 | INFO | train rank=0 | step=65450/100000 progress=65.5% loss=0.035431 lr=2.923609e-06 elapsed=24:16:37 avg_step=1.34s recent_step=1.46s eta=12:48:55 +2026-08-26 04:12:20 | INFO | train rank=0 | step=65460/100000 progress=65.5% loss=0.051575 lr=2.922105e-06 elapsed=24:16:49 avg_step=1.34s recent_step=1.20s eta=12:48:41 +2026-08-26 04:12:38 | INFO | train rank=0 | step=65470/100000 progress=65.5% loss=0.077972 lr=2.920601e-06 elapsed=24:17:08 avg_step=1.34s recent_step=1.82s eta=12:48:31 +2026-08-26 04:12:50 | INFO | train rank=0 | step=65480/100000 progress=65.5% loss=0.034401 lr=2.919097e-06 elapsed=24:17:20 avg_step=1.34s recent_step=1.20s eta=12:48:16 +2026-08-26 04:13:01 | INFO | train rank=0 | step=65490/100000 progress=65.5% loss=0.064819 lr=2.917594e-06 elapsed=24:17:31 avg_step=1.34s recent_step=1.17s eta=12:48:02 +2026-08-26 04:13:13 | INFO | train rank=0 | step=65500/100000 progress=65.5% loss=0.058624 lr=2.916091e-06 elapsed=24:17:43 avg_step=1.34s recent_step=1.18s eta=12:47:48 +2026-08-26 04:13:25 | INFO | train rank=0 | step=65510/100000 progress=65.5% loss=0.067230 lr=2.914588e-06 elapsed=24:17:55 avg_step=1.34s recent_step=1.18s eta=12:47:34 +2026-08-26 04:13:37 | INFO | train rank=0 | step=65520/100000 progress=65.5% loss=0.036636 lr=2.913085e-06 elapsed=24:18:07 avg_step=1.34s recent_step=1.19s eta=12:47:20 +2026-08-26 04:13:49 | INFO | train rank=0 | step=65530/100000 progress=65.5% loss=0.069946 lr=2.911583e-06 elapsed=24:18:19 avg_step=1.34s recent_step=1.19s eta=12:47:06 +2026-08-26 04:14:01 | INFO | train rank=0 | step=65540/100000 progress=65.5% loss=0.055252 lr=2.910080e-06 elapsed=24:18:30 avg_step=1.34s recent_step=1.19s eta=12:46:52 +2026-08-26 04:14:15 | INFO | train rank=0 | step=65550/100000 progress=65.5% loss=0.058105 lr=2.908578e-06 elapsed=24:18:45 avg_step=1.34s recent_step=1.47s eta=12:46:39 +2026-08-26 04:14:27 | INFO | train rank=0 | step=65560/100000 progress=65.6% loss=0.041542 lr=2.907077e-06 elapsed=24:18:57 avg_step=1.34s recent_step=1.21s eta=12:46:25 +2026-08-26 04:14:39 | INFO | train rank=0 | step=65570/100000 progress=65.6% loss=0.027039 lr=2.905575e-06 elapsed=24:19:09 avg_step=1.34s recent_step=1.17s eta=12:46:11 +2026-08-26 04:14:50 | INFO | train rank=0 | step=65580/100000 progress=65.6% loss=0.035439 lr=2.904074e-06 elapsed=24:19:20 avg_step=1.34s recent_step=1.15s eta=12:45:56 +2026-08-26 04:15:05 | INFO | train rank=0 | step=65590/100000 progress=65.6% loss=0.071960 lr=2.902573e-06 elapsed=24:19:35 avg_step=1.34s recent_step=1.41s eta=12:45:43 +2026-08-26 04:15:17 | INFO | train rank=0 | step=65600/100000 progress=65.6% loss=0.073975 lr=2.901072e-06 elapsed=24:19:47 avg_step=1.34s recent_step=1.25s eta=12:45:30 +2026-08-26 04:15:33 | INFO | train rank=0 | step=65610/100000 progress=65.6% loss=0.052597 lr=2.899571e-06 elapsed=24:20:03 avg_step=1.34s recent_step=1.59s eta=12:45:18 +2026-08-26 04:15:45 | INFO | train rank=0 | step=65620/100000 progress=65.6% loss=0.066925 lr=2.898071e-06 elapsed=24:20:15 avg_step=1.34s recent_step=1.17s eta=12:45:03 +2026-08-26 04:15:58 | INFO | train rank=0 | step=65630/100000 progress=65.6% loss=0.047577 lr=2.896571e-06 elapsed=24:20:28 avg_step=1.34s recent_step=1.30s eta=12:44:50 +2026-08-26 04:16:14 | INFO | train rank=0 | step=65640/100000 progress=65.6% loss=0.040474 lr=2.895071e-06 elapsed=24:20:44 avg_step=1.34s recent_step=1.64s eta=12:44:38 +2026-08-26 04:16:29 | INFO | train rank=0 | step=65650/100000 progress=65.6% loss=0.058273 lr=2.893571e-06 elapsed=24:20:59 avg_step=1.34s recent_step=1.49s eta=12:44:25 +2026-08-26 04:16:47 | INFO | train rank=0 | step=65660/100000 progress=65.7% loss=0.040634 lr=2.892072e-06 elapsed=24:21:17 avg_step=1.34s recent_step=1.82s eta=12:44:15 +2026-08-26 04:16:59 | INFO | train rank=0 | step=65670/100000 progress=65.7% loss=0.049667 lr=2.890573e-06 elapsed=24:21:29 avg_step=1.34s recent_step=1.21s eta=12:44:01 +2026-08-26 04:17:11 | INFO | train rank=0 | step=65680/100000 progress=65.7% loss=0.057739 lr=2.889074e-06 elapsed=24:21:41 avg_step=1.34s recent_step=1.16s eta=12:43:46 +2026-08-26 04:17:23 | INFO | train rank=0 | step=65690/100000 progress=65.7% loss=0.031540 lr=2.887575e-06 elapsed=24:21:53 avg_step=1.34s recent_step=1.19s eta=12:43:32 +2026-08-26 04:17:36 | INFO | train rank=0 | step=65700/100000 progress=65.7% loss=0.048607 lr=2.886076e-06 elapsed=24:22:06 avg_step=1.34s recent_step=1.36s eta=12:43:19 +2026-08-26 04:17:50 | INFO | train rank=0 | step=65710/100000 progress=65.7% loss=0.042404 lr=2.884578e-06 elapsed=24:22:20 avg_step=1.34s recent_step=1.41s eta=12:43:06 +2026-08-26 04:18:02 | INFO | train rank=0 | step=65720/100000 progress=65.7% loss=0.048340 lr=2.883080e-06 elapsed=24:22:32 avg_step=1.34s recent_step=1.19s eta=12:42:52 +2026-08-26 04:18:14 | INFO | train rank=0 | step=65730/100000 progress=65.7% loss=0.055908 lr=2.881582e-06 elapsed=24:22:44 avg_step=1.34s recent_step=1.14s eta=12:42:38 +2026-08-26 04:18:31 | INFO | train rank=0 | step=65740/100000 progress=65.7% loss=0.033615 lr=2.880084e-06 elapsed=24:23:01 avg_step=1.34s recent_step=1.73s eta=12:42:26 +2026-08-26 04:18:46 | INFO | train rank=0 | step=65750/100000 progress=65.8% loss=0.037827 lr=2.878587e-06 elapsed=24:23:16 avg_step=1.34s recent_step=1.50s eta=12:42:14 +2026-08-26 04:19:00 | INFO | train rank=0 | step=65760/100000 progress=65.8% loss=0.041550 lr=2.877090e-06 elapsed=24:23:30 avg_step=1.34s recent_step=1.39s eta=12:42:01 +2026-08-26 04:19:12 | INFO | train rank=0 | step=65770/100000 progress=65.8% loss=0.045258 lr=2.875593e-06 elapsed=24:23:42 avg_step=1.34s recent_step=1.22s eta=12:41:47 +2026-08-26 04:19:24 | INFO | train rank=0 | step=65780/100000 progress=65.8% loss=0.038696 lr=2.874096e-06 elapsed=24:23:54 avg_step=1.34s recent_step=1.18s eta=12:41:33 +2026-08-26 04:19:36 | INFO | train rank=0 | step=65790/100000 progress=65.8% loss=0.047546 lr=2.872600e-06 elapsed=24:24:06 avg_step=1.34s recent_step=1.21s eta=12:41:19 +2026-08-26 04:19:52 | INFO | train rank=0 | step=65800/100000 progress=65.8% loss=0.049057 lr=2.871104e-06 elapsed=24:24:21 avg_step=1.34s recent_step=1.56s eta=12:41:06 +2026-08-26 04:20:07 | INFO | train rank=0 | step=65810/100000 progress=65.8% loss=0.054863 lr=2.869608e-06 elapsed=24:24:37 avg_step=1.34s recent_step=1.52s eta=12:40:54 +2026-08-26 04:20:19 | INFO | train rank=0 | step=65820/100000 progress=65.8% loss=0.025841 lr=2.868112e-06 elapsed=24:24:49 avg_step=1.34s recent_step=1.20s eta=12:40:40 +2026-08-26 04:20:34 | INFO | train rank=0 | step=65830/100000 progress=65.8% loss=0.042191 lr=2.866616e-06 elapsed=24:25:04 avg_step=1.34s recent_step=1.49s eta=12:40:27 +2026-08-26 04:20:45 | INFO | train rank=0 | step=65840/100000 progress=65.8% loss=0.046570 lr=2.865121e-06 elapsed=24:25:15 avg_step=1.34s recent_step=1.15s eta=12:40:13 +2026-08-26 04:21:00 | INFO | train rank=0 | step=65850/100000 progress=65.8% loss=0.043121 lr=2.863626e-06 elapsed=24:25:30 avg_step=1.34s recent_step=1.47s eta=12:40:00 +2026-08-26 04:21:11 | INFO | train rank=0 | step=65860/100000 progress=65.9% loss=0.033546 lr=2.862131e-06 elapsed=24:25:41 avg_step=1.34s recent_step=1.12s eta=12:39:46 +2026-08-26 04:21:29 | INFO | train rank=0 | step=65870/100000 progress=65.9% loss=0.054596 lr=2.860637e-06 elapsed=24:25:59 avg_step=1.34s recent_step=1.83s eta=12:39:35 +2026-08-26 04:21:41 | INFO | train rank=0 | step=65880/100000 progress=65.9% loss=0.039108 lr=2.859142e-06 elapsed=24:26:11 avg_step=1.34s recent_step=1.20s eta=12:39:21 +2026-08-26 04:21:53 | INFO | train rank=0 | step=65890/100000 progress=65.9% loss=0.035873 lr=2.857648e-06 elapsed=24:26:23 avg_step=1.34s recent_step=1.15s eta=12:39:07 +2026-08-26 04:22:05 | INFO | train rank=0 | step=65900/100000 progress=65.9% loss=0.033203 lr=2.856154e-06 elapsed=24:26:34 avg_step=1.34s recent_step=1.17s eta=12:38:53 +2026-08-26 04:22:19 | INFO | train rank=0 | step=65910/100000 progress=65.9% loss=0.049896 lr=2.854661e-06 elapsed=24:26:49 avg_step=1.34s recent_step=1.41s eta=12:38:40 +2026-08-26 04:22:31 | INFO | train rank=0 | step=65920/100000 progress=65.9% loss=0.045845 lr=2.853167e-06 elapsed=24:27:00 avg_step=1.34s recent_step=1.19s eta=12:38:25 +2026-08-26 04:22:45 | INFO | train rank=0 | step=65930/100000 progress=65.9% loss=0.058426 lr=2.851674e-06 elapsed=24:27:14 avg_step=1.34s recent_step=1.41s eta=12:38:12 +2026-08-26 04:22:58 | INFO | train rank=0 | step=65940/100000 progress=65.9% loss=0.045395 lr=2.850181e-06 elapsed=24:27:27 avg_step=1.34s recent_step=1.30s eta=12:37:59 +2026-08-26 04:23:13 | INFO | train rank=0 | step=65950/100000 progress=66.0% loss=0.044586 lr=2.848688e-06 elapsed=24:27:43 avg_step=1.34s recent_step=1.57s eta=12:37:47 +2026-08-26 04:23:25 | INFO | train rank=0 | step=65960/100000 progress=66.0% loss=0.046921 lr=2.847196e-06 elapsed=24:27:55 avg_step=1.34s recent_step=1.16s eta=12:37:33 +2026-08-26 04:23:39 | INFO | train rank=0 | step=65970/100000 progress=66.0% loss=0.048386 lr=2.845704e-06 elapsed=24:28:09 avg_step=1.34s recent_step=1.41s eta=12:37:20 +2026-08-26 04:23:51 | INFO | train rank=0 | step=65980/100000 progress=66.0% loss=0.030785 lr=2.844212e-06 elapsed=24:28:21 avg_step=1.34s recent_step=1.22s eta=12:37:06 +2026-08-26 04:24:07 | INFO | train rank=0 | step=65990/100000 progress=66.0% loss=0.036453 lr=2.842720e-06 elapsed=24:28:37 avg_step=1.34s recent_step=1.59s eta=12:36:54 +2026-08-26 04:24:20 | INFO | train rank=0 | step=66000/100000 progress=66.0% loss=0.054382 lr=2.841228e-06 elapsed=24:28:49 avg_step=1.34s recent_step=1.25s eta=12:36:40 +2026-08-26 04:24:35 | INFO | train rank=0 | step=66010/100000 progress=66.0% loss=0.058411 lr=2.839737e-06 elapsed=24:29:05 avg_step=1.34s recent_step=1.55s eta=12:36:28 +2026-08-26 04:24:47 | INFO | train rank=0 | step=66020/100000 progress=66.0% loss=0.040710 lr=2.838246e-06 elapsed=24:29:16 avg_step=1.34s recent_step=1.15s eta=12:36:13 +2026-08-26 04:25:00 | INFO | train rank=0 | step=66030/100000 progress=66.0% loss=0.052826 lr=2.836755e-06 elapsed=24:29:30 avg_step=1.34s recent_step=1.38s eta=12:36:00 +2026-08-26 04:25:12 | INFO | train rank=0 | step=66040/100000 progress=66.0% loss=0.046494 lr=2.835264e-06 elapsed=24:29:42 avg_step=1.34s recent_step=1.15s eta=12:35:46 +2026-08-26 04:25:25 | INFO | train rank=0 | step=66050/100000 progress=66.0% loss=0.068146 lr=2.833774e-06 elapsed=24:29:55 avg_step=1.34s recent_step=1.32s eta=12:35:32 +2026-08-26 04:25:41 | INFO | train rank=0 | step=66060/100000 progress=66.1% loss=0.024990 lr=2.832284e-06 elapsed=24:30:11 avg_step=1.34s recent_step=1.63s eta=12:35:21 +2026-08-26 04:25:53 | INFO | train rank=0 | step=66070/100000 progress=66.1% loss=0.043320 lr=2.830794e-06 elapsed=24:30:23 avg_step=1.34s recent_step=1.16s eta=12:35:06 +2026-08-26 04:26:05 | INFO | train rank=0 | step=66080/100000 progress=66.1% loss=0.034851 lr=2.829304e-06 elapsed=24:30:35 avg_step=1.34s recent_step=1.19s eta=12:34:52 +2026-08-26 04:26:20 | INFO | train rank=0 | step=66090/100000 progress=66.1% loss=0.033775 lr=2.827815e-06 elapsed=24:30:49 avg_step=1.34s recent_step=1.46s eta=12:34:39 +2026-08-26 04:26:31 | INFO | train rank=0 | step=66100/100000 progress=66.1% loss=0.046921 lr=2.826326e-06 elapsed=24:31:01 avg_step=1.34s recent_step=1.17s eta=12:34:25 +2026-08-26 04:26:46 | INFO | train rank=0 | step=66110/100000 progress=66.1% loss=0.079865 lr=2.824837e-06 elapsed=24:31:16 avg_step=1.34s recent_step=1.51s eta=12:34:13 +2026-08-26 04:26:58 | INFO | train rank=0 | step=66120/100000 progress=66.1% loss=0.047974 lr=2.823348e-06 elapsed=24:31:27 avg_step=1.34s recent_step=1.12s eta=12:33:58 +2026-08-26 04:27:11 | INFO | train rank=0 | step=66130/100000 progress=66.1% loss=0.072266 lr=2.821860e-06 elapsed=24:31:41 avg_step=1.34s recent_step=1.38s eta=12:33:45 +2026-08-26 04:27:23 | INFO | train rank=0 | step=66140/100000 progress=66.1% loss=0.055557 lr=2.820372e-06 elapsed=24:31:53 avg_step=1.34s recent_step=1.19s eta=12:33:31 +2026-08-26 04:27:45 | INFO | train rank=0 | step=66150/100000 progress=66.1% loss=0.052094 lr=2.818884e-06 elapsed=24:32:14 avg_step=1.34s recent_step=2.13s eta=12:33:22 +2026-08-26 04:27:56 | INFO | train rank=0 | step=66160/100000 progress=66.2% loss=0.049042 lr=2.817396e-06 elapsed=24:32:26 avg_step=1.34s recent_step=1.12s eta=12:33:07 +2026-08-26 04:28:08 | INFO | train rank=0 | step=66170/100000 progress=66.2% loss=0.052673 lr=2.815908e-06 elapsed=24:32:38 avg_step=1.34s recent_step=1.24s eta=12:32:54 +2026-08-26 04:28:20 | INFO | train rank=0 | step=66180/100000 progress=66.2% loss=0.051529 lr=2.814421e-06 elapsed=24:32:50 avg_step=1.34s recent_step=1.15s eta=12:32:39 +2026-08-26 04:28:36 | INFO | train rank=0 | step=66190/100000 progress=66.2% loss=0.034851 lr=2.812934e-06 elapsed=24:33:06 avg_step=1.34s recent_step=1.64s eta=12:32:28 +2026-08-26 04:28:49 | INFO | train rank=0 | step=66200/100000 progress=66.2% loss=0.056915 lr=2.811447e-06 elapsed=24:33:19 avg_step=1.34s recent_step=1.27s eta=12:32:14 +2026-08-26 04:29:02 | INFO | train rank=0 | step=66210/100000 progress=66.2% loss=0.043396 lr=2.809961e-06 elapsed=24:33:31 avg_step=1.34s recent_step=1.27s eta=12:32:00 +2026-08-26 04:29:14 | INFO | train rank=0 | step=66220/100000 progress=66.2% loss=0.039307 lr=2.808475e-06 elapsed=24:33:44 avg_step=1.34s recent_step=1.27s eta=12:31:46 +2026-08-26 04:29:34 | INFO | train rank=0 | step=66230/100000 progress=66.2% loss=0.047134 lr=2.806988e-06 elapsed=24:34:04 avg_step=1.34s recent_step=2.00s eta=12:31:37 +2026-08-26 04:29:46 | INFO | train rank=0 | step=66240/100000 progress=66.2% loss=0.036346 lr=2.805503e-06 elapsed=24:34:16 avg_step=1.34s recent_step=1.17s eta=12:31:22 +2026-08-26 04:29:58 | INFO | train rank=0 | step=66250/100000 progress=66.2% loss=0.049118 lr=2.804017e-06 elapsed=24:34:27 avg_step=1.34s recent_step=1.16s eta=12:31:08 +2026-08-26 04:30:12 | INFO | train rank=0 | step=66260/100000 progress=66.3% loss=0.049465 lr=2.802532e-06 elapsed=24:34:42 avg_step=1.34s recent_step=1.41s eta=12:30:55 +2026-08-26 04:30:25 | INFO | train rank=0 | step=66270/100000 progress=66.3% loss=0.038780 lr=2.801047e-06 elapsed=24:34:55 avg_step=1.34s recent_step=1.35s eta=12:30:42 +2026-08-26 04:30:41 | INFO | train rank=0 | step=66280/100000 progress=66.3% loss=0.054413 lr=2.799562e-06 elapsed=24:35:11 avg_step=1.34s recent_step=1.58s eta=12:30:30 +2026-08-26 04:30:53 | INFO | train rank=0 | step=66290/100000 progress=66.3% loss=0.115479 lr=2.798077e-06 elapsed=24:35:23 avg_step=1.34s recent_step=1.22s eta=12:30:16 +2026-08-26 04:31:10 | INFO | train rank=0 | step=66300/100000 progress=66.3% loss=0.060242 lr=2.796593e-06 elapsed=24:35:40 avg_step=1.34s recent_step=1.72s eta=12:30:04 +2026-08-26 04:31:22 | INFO | train rank=0 | step=66310/100000 progress=66.3% loss=0.059738 lr=2.795109e-06 elapsed=24:35:52 avg_step=1.34s recent_step=1.16s eta=12:29:50 +2026-08-26 04:31:34 | INFO | train rank=0 | step=66320/100000 progress=66.3% loss=0.052078 lr=2.793625e-06 elapsed=24:36:04 avg_step=1.34s recent_step=1.19s eta=12:29:36 +2026-08-26 04:31:45 | INFO | train rank=0 | step=66330/100000 progress=66.3% loss=0.035690 lr=2.792141e-06 elapsed=24:36:15 avg_step=1.34s recent_step=1.10s eta=12:29:21 +2026-08-26 04:31:57 | INFO | train rank=0 | step=66340/100000 progress=66.3% loss=0.048080 lr=2.790658e-06 elapsed=24:36:27 avg_step=1.34s recent_step=1.18s eta=12:29:07 +2026-08-26 04:32:10 | INFO | train rank=0 | step=66350/100000 progress=66.3% loss=0.031166 lr=2.789174e-06 elapsed=24:36:39 avg_step=1.34s recent_step=1.29s eta=12:28:54 +2026-08-26 04:32:26 | INFO | train rank=0 | step=66360/100000 progress=66.4% loss=0.061447 lr=2.787692e-06 elapsed=24:36:55 avg_step=1.34s recent_step=1.59s eta=12:28:42 +2026-08-26 04:32:40 | INFO | train rank=0 | step=66370/100000 progress=66.4% loss=0.059921 lr=2.786209e-06 elapsed=24:37:10 avg_step=1.34s recent_step=1.49s eta=12:28:29 +2026-08-26 04:32:52 | INFO | train rank=0 | step=66380/100000 progress=66.4% loss=0.038055 lr=2.784726e-06 elapsed=24:37:22 avg_step=1.34s recent_step=1.14s eta=12:28:15 +2026-08-26 04:33:09 | INFO | train rank=0 | step=66390/100000 progress=66.4% loss=0.060150 lr=2.783244e-06 elapsed=24:37:39 avg_step=1.34s recent_step=1.72s eta=12:28:03 +2026-08-26 04:33:24 | INFO | train rank=0 | step=66400/100000 progress=66.4% loss=0.046844 lr=2.781762e-06 elapsed=24:37:54 avg_step=1.34s recent_step=1.50s eta=12:27:51 +2026-08-26 04:33:35 | INFO | train rank=0 | step=66410/100000 progress=66.4% loss=0.034744 lr=2.780281e-06 elapsed=24:38:05 avg_step=1.34s recent_step=1.13s eta=12:27:37 +2026-08-26 04:33:50 | INFO | train rank=0 | step=66420/100000 progress=66.4% loss=0.045990 lr=2.778799e-06 elapsed=24:38:19 avg_step=1.34s recent_step=1.42s eta=12:27:24 +2026-08-26 04:34:06 | INFO | train rank=0 | step=66430/100000 progress=66.4% loss=0.061783 lr=2.777318e-06 elapsed=24:38:36 avg_step=1.34s recent_step=1.66s eta=12:27:12 +2026-08-26 04:34:17 | INFO | train rank=0 | step=66440/100000 progress=66.4% loss=0.039246 lr=2.775837e-06 elapsed=24:38:47 avg_step=1.34s recent_step=1.11s eta=12:26:57 +2026-08-26 04:34:29 | INFO | train rank=0 | step=66450/100000 progress=66.5% loss=0.048630 lr=2.774356e-06 elapsed=24:38:59 avg_step=1.34s recent_step=1.15s eta=12:26:43 +2026-08-26 04:34:43 | INFO | train rank=0 | step=66460/100000 progress=66.5% loss=0.051514 lr=2.772876e-06 elapsed=24:39:13 avg_step=1.34s recent_step=1.46s eta=12:26:30 +2026-08-26 04:34:55 | INFO | train rank=0 | step=66470/100000 progress=66.5% loss=0.060364 lr=2.771395e-06 elapsed=24:39:25 avg_step=1.34s recent_step=1.16s eta=12:26:16 +2026-08-26 04:35:08 | INFO | train rank=0 | step=66480/100000 progress=66.5% loss=0.041504 lr=2.769915e-06 elapsed=24:39:38 avg_step=1.34s recent_step=1.32s eta=12:26:03 +2026-08-26 04:35:20 | INFO | train rank=0 | step=66490/100000 progress=66.5% loss=0.070038 lr=2.768435e-06 elapsed=24:39:50 avg_step=1.34s recent_step=1.20s eta=12:25:49 +2026-08-26 04:35:32 | INFO | train rank=0 | step=66500/100000 progress=66.5% loss=0.048576 lr=2.766956e-06 elapsed=24:40:01 avg_step=1.34s recent_step=1.15s eta=12:25:34 +2026-08-26 04:35:48 | INFO | train rank=0 | step=66510/100000 progress=66.5% loss=0.053802 lr=2.765477e-06 elapsed=24:40:18 avg_step=1.34s recent_step=1.68s eta=12:25:23 +2026-08-26 04:36:01 | INFO | train rank=0 | step=66520/100000 progress=66.5% loss=0.066513 lr=2.763998e-06 elapsed=24:40:31 avg_step=1.34s recent_step=1.23s eta=12:25:09 +2026-08-26 04:36:12 | INFO | train rank=0 | step=66530/100000 progress=66.5% loss=0.058014 lr=2.762519e-06 elapsed=24:40:42 avg_step=1.34s recent_step=1.15s eta=12:24:54 +2026-08-26 04:36:26 | INFO | train rank=0 | step=66540/100000 progress=66.5% loss=0.081329 lr=2.761040e-06 elapsed=24:40:56 avg_step=1.34s recent_step=1.40s eta=12:24:41 +2026-08-26 04:36:39 | INFO | train rank=0 | step=66550/100000 progress=66.5% loss=0.044418 lr=2.759562e-06 elapsed=24:41:09 avg_step=1.34s recent_step=1.28s eta=12:24:28 +2026-08-26 04:36:50 | INFO | train rank=0 | step=66560/100000 progress=66.6% loss=0.052490 lr=2.758084e-06 elapsed=24:41:20 avg_step=1.34s recent_step=1.14s eta=12:24:13 +2026-08-26 04:37:04 | INFO | train rank=0 | step=66570/100000 progress=66.6% loss=0.061157 lr=2.756606e-06 elapsed=24:41:34 avg_step=1.34s recent_step=1.38s eta=12:24:00 +2026-08-26 04:37:19 | INFO | train rank=0 | step=66580/100000 progress=66.6% loss=0.037636 lr=2.755129e-06 elapsed=24:41:49 avg_step=1.34s recent_step=1.49s eta=12:23:48 +2026-08-26 04:37:32 | INFO | train rank=0 | step=66590/100000 progress=66.6% loss=0.030289 lr=2.753651e-06 elapsed=24:42:02 avg_step=1.34s recent_step=1.29s eta=12:23:34 +2026-08-26 04:37:47 | INFO | train rank=0 | step=66600/100000 progress=66.6% loss=0.050125 lr=2.752174e-06 elapsed=24:42:16 avg_step=1.34s recent_step=1.47s eta=12:23:22 +2026-08-26 04:37:59 | INFO | train rank=0 | step=66610/100000 progress=66.6% loss=0.089264 lr=2.750697e-06 elapsed=24:42:29 avg_step=1.34s recent_step=1.24s eta=12:23:08 +2026-08-26 04:38:12 | INFO | train rank=0 | step=66620/100000 progress=66.6% loss=0.069275 lr=2.749221e-06 elapsed=24:42:42 avg_step=1.34s recent_step=1.32s eta=12:22:54 +2026-08-26 04:38:26 | INFO | train rank=0 | step=66630/100000 progress=66.6% loss=0.028633 lr=2.747744e-06 elapsed=24:42:56 avg_step=1.34s recent_step=1.35s eta=12:22:41 +2026-08-26 04:38:37 | INFO | train rank=0 | step=66640/100000 progress=66.6% loss=0.057556 lr=2.746268e-06 elapsed=24:43:07 avg_step=1.34s recent_step=1.17s eta=12:22:27 +2026-08-26 04:38:49 | INFO | train rank=0 | step=66650/100000 progress=66.6% loss=0.044403 lr=2.744792e-06 elapsed=24:43:19 avg_step=1.34s recent_step=1.13s eta=12:22:12 +2026-08-26 04:39:04 | INFO | train rank=0 | step=66660/100000 progress=66.7% loss=0.029312 lr=2.743317e-06 elapsed=24:43:34 avg_step=1.34s recent_step=1.53s eta=12:22:00 +2026-08-26 04:39:20 | INFO | train rank=0 | step=66670/100000 progress=66.7% loss=0.038818 lr=2.741841e-06 elapsed=24:43:50 avg_step=1.34s recent_step=1.64s eta=12:21:48 +2026-08-26 04:39:33 | INFO | train rank=0 | step=66680/100000 progress=66.7% loss=0.073914 lr=2.740366e-06 elapsed=24:44:03 avg_step=1.34s recent_step=1.30s eta=12:21:35 +2026-08-26 04:39:47 | INFO | train rank=0 | step=66690/100000 progress=66.7% loss=0.038757 lr=2.738891e-06 elapsed=24:44:17 avg_step=1.34s recent_step=1.39s eta=12:21:22 +2026-08-26 04:40:00 | INFO | train rank=0 | step=66700/100000 progress=66.7% loss=0.046036 lr=2.737417e-06 elapsed=24:44:30 avg_step=1.34s recent_step=1.32s eta=12:21:08 +2026-08-26 04:40:13 | INFO | train rank=0 | step=66710/100000 progress=66.7% loss=0.056778 lr=2.735942e-06 elapsed=24:44:43 avg_step=1.34s recent_step=1.22s eta=12:20:54 +2026-08-26 04:40:24 | INFO | train rank=0 | step=66720/100000 progress=66.7% loss=0.036209 lr=2.734468e-06 elapsed=24:44:54 avg_step=1.34s recent_step=1.13s eta=12:20:40 +2026-08-26 04:40:35 | INFO | train rank=0 | step=66730/100000 progress=66.7% loss=0.034538 lr=2.732995e-06 elapsed=24:45:05 avg_step=1.34s recent_step=1.14s eta=12:20:26 +2026-08-26 04:40:50 | INFO | train rank=0 | step=66740/100000 progress=66.7% loss=0.033890 lr=2.731521e-06 elapsed=24:45:20 avg_step=1.34s recent_step=1.51s eta=12:20:13 +2026-08-26 04:41:05 | INFO | train rank=0 | step=66750/100000 progress=66.8% loss=0.048767 lr=2.730048e-06 elapsed=24:45:35 avg_step=1.34s recent_step=1.48s eta=12:20:00 +2026-08-26 04:41:17 | INFO | train rank=0 | step=66760/100000 progress=66.8% loss=0.060104 lr=2.728574e-06 elapsed=24:45:47 avg_step=1.34s recent_step=1.14s eta=12:19:46 +2026-08-26 04:41:35 | INFO | train rank=0 | step=66770/100000 progress=66.8% loss=0.027931 lr=2.727101e-06 elapsed=24:46:05 avg_step=1.34s recent_step=1.79s eta=12:19:35 +2026-08-26 04:41:46 | INFO | train rank=0 | step=66780/100000 progress=66.8% loss=0.064362 lr=2.725629e-06 elapsed=24:46:16 avg_step=1.34s recent_step=1.16s eta=12:19:21 +2026-08-26 04:42:01 | INFO | train rank=0 | step=66790/100000 progress=66.8% loss=0.031311 lr=2.724156e-06 elapsed=24:46:30 avg_step=1.34s recent_step=1.43s eta=12:19:08 +2026-08-26 04:42:12 | INFO | train rank=0 | step=66800/100000 progress=66.8% loss=0.046448 lr=2.722684e-06 elapsed=24:46:42 avg_step=1.34s recent_step=1.12s eta=12:18:54 +2026-08-26 04:42:27 | INFO | train rank=0 | step=66810/100000 progress=66.8% loss=0.068115 lr=2.721212e-06 elapsed=24:46:57 avg_step=1.34s recent_step=1.56s eta=12:18:41 +2026-08-26 04:42:40 | INFO | train rank=0 | step=66820/100000 progress=66.8% loss=0.038956 lr=2.719741e-06 elapsed=24:47:10 avg_step=1.34s recent_step=1.31s eta=12:18:28 +2026-08-26 04:42:56 | INFO | train rank=0 | step=66830/100000 progress=66.8% loss=0.040527 lr=2.718269e-06 elapsed=24:47:26 avg_step=1.34s recent_step=1.55s eta=12:18:15 +2026-08-26 04:43:08 | INFO | train rank=0 | step=66840/100000 progress=66.8% loss=0.031647 lr=2.716798e-06 elapsed=24:47:37 avg_step=1.34s recent_step=1.16s eta=12:18:01 +2026-08-26 04:43:21 | INFO | train rank=0 | step=66850/100000 progress=66.8% loss=0.040955 lr=2.715327e-06 elapsed=24:47:51 avg_step=1.34s recent_step=1.36s eta=12:17:48 +2026-08-26 04:43:38 | INFO | train rank=0 | step=66860/100000 progress=66.9% loss=0.050903 lr=2.713857e-06 elapsed=24:48:08 avg_step=1.34s recent_step=1.70s eta=12:17:37 +2026-08-26 04:43:50 | INFO | train rank=0 | step=66870/100000 progress=66.9% loss=0.057373 lr=2.712386e-06 elapsed=24:48:20 avg_step=1.34s recent_step=1.19s eta=12:17:22 +2026-08-26 04:44:02 | INFO | train rank=0 | step=66880/100000 progress=66.9% loss=0.034821 lr=2.710916e-06 elapsed=24:48:32 avg_step=1.34s recent_step=1.17s eta=12:17:08 +2026-08-26 04:44:14 | INFO | train rank=0 | step=66890/100000 progress=66.9% loss=0.029648 lr=2.709446e-06 elapsed=24:48:44 avg_step=1.34s recent_step=1.19s eta=12:16:54 +2026-08-26 04:44:30 | INFO | train rank=0 | step=66900/100000 progress=66.9% loss=0.041931 lr=2.707977e-06 elapsed=24:49:00 avg_step=1.34s recent_step=1.63s eta=12:16:42 +2026-08-26 04:44:41 | INFO | train rank=0 | step=66910/100000 progress=66.9% loss=0.063721 lr=2.706507e-06 elapsed=24:49:11 avg_step=1.34s recent_step=1.12s eta=12:16:28 +2026-08-26 04:44:55 | INFO | train rank=0 | step=66920/100000 progress=66.9% loss=0.054382 lr=2.705038e-06 elapsed=24:49:25 avg_step=1.34s recent_step=1.42s eta=12:16:15 +2026-08-26 04:45:08 | INFO | train rank=0 | step=66930/100000 progress=66.9% loss=0.065094 lr=2.703569e-06 elapsed=24:49:38 avg_step=1.34s recent_step=1.23s eta=12:16:01 +2026-08-26 04:45:22 | INFO | train rank=0 | step=66940/100000 progress=66.9% loss=0.046097 lr=2.702101e-06 elapsed=24:49:52 avg_step=1.34s recent_step=1.39s eta=12:15:48 +2026-08-26 04:45:36 | INFO | train rank=0 | step=66950/100000 progress=67.0% loss=0.065155 lr=2.700632e-06 elapsed=24:50:06 avg_step=1.34s recent_step=1.44s eta=12:15:35 +2026-08-26 04:45:48 | INFO | train rank=0 | step=66960/100000 progress=67.0% loss=0.048309 lr=2.699164e-06 elapsed=24:50:18 avg_step=1.34s recent_step=1.19s eta=12:15:21 +2026-08-26 04:46:00 | INFO | train rank=0 | step=66970/100000 progress=67.0% loss=0.057449 lr=2.697696e-06 elapsed=24:50:30 avg_step=1.34s recent_step=1.17s eta=12:15:07 +2026-08-26 04:46:21 | INFO | train rank=0 | step=66980/100000 progress=67.0% loss=0.044388 lr=2.696229e-06 elapsed=24:50:51 avg_step=1.34s recent_step=2.12s eta=12:14:57 +2026-08-26 04:46:32 | INFO | train rank=0 | step=66990/100000 progress=67.0% loss=0.050842 lr=2.694761e-06 elapsed=24:51:02 avg_step=1.34s recent_step=1.11s eta=12:14:43 +2026-08-26 04:46:44 | INFO | train rank=0 | step=67000/100000 progress=67.0% loss=0.059525 lr=2.693294e-06 elapsed=24:51:14 avg_step=1.34s recent_step=1.19s eta=12:14:29 +2026-08-26 04:46:56 | INFO | train rank=0 | step=67010/100000 progress=67.0% loss=0.043655 lr=2.691827e-06 elapsed=24:51:26 avg_step=1.34s recent_step=1.18s eta=12:14:15 +2026-08-26 04:47:13 | INFO | train rank=0 | step=67020/100000 progress=67.0% loss=0.055023 lr=2.690361e-06 elapsed=24:51:43 avg_step=1.34s recent_step=1.76s eta=12:14:04 +2026-08-26 04:47:26 | INFO | train rank=0 | step=67030/100000 progress=67.0% loss=0.030365 lr=2.688894e-06 elapsed=24:51:56 avg_step=1.34s recent_step=1.25s eta=12:13:50 +2026-08-26 04:47:37 | INFO | train rank=0 | step=67040/100000 progress=67.0% loss=0.037117 lr=2.687428e-06 elapsed=24:52:07 avg_step=1.34s recent_step=1.13s eta=12:13:35 +2026-08-26 04:47:49 | INFO | train rank=0 | step=67050/100000 progress=67.0% loss=0.034866 lr=2.685962e-06 elapsed=24:52:19 avg_step=1.34s recent_step=1.20s eta=12:13:21 +2026-08-26 04:48:07 | INFO | train rank=0 | step=67060/100000 progress=67.1% loss=0.053711 lr=2.684497e-06 elapsed=24:52:37 avg_step=1.34s recent_step=1.85s eta=12:13:11 +2026-08-26 04:48:21 | INFO | train rank=0 | step=67070/100000 progress=67.1% loss=0.041138 lr=2.683031e-06 elapsed=24:52:51 avg_step=1.34s recent_step=1.38s eta=12:12:57 +2026-08-26 04:48:37 | INFO | train rank=0 | step=67080/100000 progress=67.1% loss=0.037769 lr=2.681566e-06 elapsed=24:53:07 avg_step=1.34s recent_step=1.55s eta=12:12:45 +2026-08-26 04:48:50 | INFO | train rank=0 | step=67090/100000 progress=67.1% loss=0.044388 lr=2.680101e-06 elapsed=24:53:20 avg_step=1.34s recent_step=1.29s eta=12:12:32 +2026-08-26 04:49:03 | INFO | train rank=0 | step=67100/100000 progress=67.1% loss=0.063599 lr=2.678637e-06 elapsed=24:53:33 avg_step=1.34s recent_step=1.37s eta=12:12:18 +2026-08-26 04:49:15 | INFO | train rank=0 | step=67110/100000 progress=67.1% loss=0.057220 lr=2.677173e-06 elapsed=24:53:45 avg_step=1.34s recent_step=1.20s eta=12:12:04 +2026-08-26 04:49:32 | INFO | train rank=0 | step=67120/100000 progress=67.1% loss=0.041840 lr=2.675708e-06 elapsed=24:54:02 avg_step=1.34s recent_step=1.68s eta=12:11:53 +2026-08-26 04:49:44 | INFO | train rank=0 | step=67130/100000 progress=67.1% loss=0.044037 lr=2.674245e-06 elapsed=24:54:14 avg_step=1.34s recent_step=1.17s eta=12:11:38 +2026-08-26 04:49:56 | INFO | train rank=0 | step=67140/100000 progress=67.1% loss=0.030510 lr=2.672781e-06 elapsed=24:54:26 avg_step=1.34s recent_step=1.22s eta=12:11:24 +2026-08-26 04:50:07 | INFO | train rank=0 | step=67150/100000 progress=67.2% loss=0.043671 lr=2.671318e-06 elapsed=24:54:37 avg_step=1.34s recent_step=1.13s eta=12:11:10 +2026-08-26 04:50:22 | INFO | train rank=0 | step=67160/100000 progress=67.2% loss=0.053467 lr=2.669855e-06 elapsed=24:54:52 avg_step=1.34s recent_step=1.46s eta=12:10:57 +2026-08-26 04:50:34 | INFO | train rank=0 | step=67170/100000 progress=67.2% loss=0.057770 lr=2.668392e-06 elapsed=24:55:04 avg_step=1.34s recent_step=1.20s eta=12:10:43 +2026-08-26 04:50:50 | INFO | train rank=0 | step=67180/100000 progress=67.2% loss=0.030144 lr=2.666929e-06 elapsed=24:55:20 avg_step=1.34s recent_step=1.62s eta=12:10:31 +2026-08-26 04:51:02 | INFO | train rank=0 | step=67190/100000 progress=67.2% loss=0.038696 lr=2.665467e-06 elapsed=24:55:32 avg_step=1.34s recent_step=1.18s eta=12:10:17 +2026-08-26 04:51:16 | INFO | train rank=0 | step=67200/100000 progress=67.2% loss=0.049835 lr=2.664005e-06 elapsed=24:55:46 avg_step=1.34s recent_step=1.38s eta=12:10:04 +2026-08-26 04:51:28 | INFO | train rank=0 | step=67210/100000 progress=67.2% loss=0.057129 lr=2.662543e-06 elapsed=24:55:58 avg_step=1.34s recent_step=1.23s eta=12:09:50 +2026-08-26 04:51:40 | INFO | train rank=0 | step=67220/100000 progress=67.2% loss=0.043961 lr=2.661082e-06 elapsed=24:56:09 avg_step=1.34s recent_step=1.13s eta=12:09:36 +2026-08-26 04:51:51 | INFO | train rank=0 | step=67230/100000 progress=67.2% loss=0.047607 lr=2.659620e-06 elapsed=24:56:21 avg_step=1.34s recent_step=1.14s eta=12:09:22 +2026-08-26 04:52:09 | INFO | train rank=0 | step=67240/100000 progress=67.2% loss=0.054398 lr=2.658159e-06 elapsed=24:56:38 avg_step=1.34s recent_step=1.76s eta=12:09:10 +2026-08-26 04:52:23 | INFO | train rank=0 | step=67250/100000 progress=67.2% loss=0.034103 lr=2.656699e-06 elapsed=24:56:53 avg_step=1.34s recent_step=1.42s eta=12:08:58 +2026-08-26 04:52:38 | INFO | train rank=0 | step=67260/100000 progress=67.3% loss=0.040741 lr=2.655238e-06 elapsed=24:57:08 avg_step=1.34s recent_step=1.55s eta=12:08:45 +2026-08-26 04:52:50 | INFO | train rank=0 | step=67270/100000 progress=67.3% loss=0.042114 lr=2.653778e-06 elapsed=24:57:20 avg_step=1.34s recent_step=1.17s eta=12:08:31 +2026-08-26 04:53:05 | INFO | train rank=0 | step=67280/100000 progress=67.3% loss=0.039627 lr=2.652318e-06 elapsed=24:57:35 avg_step=1.34s recent_step=1.50s eta=12:08:18 +2026-08-26 04:53:19 | INFO | train rank=0 | step=67290/100000 progress=67.3% loss=0.043503 lr=2.650858e-06 elapsed=24:57:49 avg_step=1.34s recent_step=1.39s eta=12:08:05 +2026-08-26 04:53:33 | INFO | train rank=0 | step=67300/100000 progress=67.3% loss=0.036343 lr=2.649399e-06 elapsed=24:58:03 avg_step=1.34s recent_step=1.41s eta=12:07:52 +2026-08-26 04:53:47 | INFO | train rank=0 | step=67310/100000 progress=67.3% loss=0.041317 lr=2.647939e-06 elapsed=24:58:17 avg_step=1.34s recent_step=1.45s eta=12:07:40 +2026-08-26 04:54:02 | INFO | train rank=0 | step=67320/100000 progress=67.3% loss=0.033295 lr=2.646480e-06 elapsed=24:58:31 avg_step=1.34s recent_step=1.42s eta=12:07:27 +2026-08-26 04:54:13 | INFO | train rank=0 | step=67330/100000 progress=67.3% loss=0.052979 lr=2.645022e-06 elapsed=24:58:43 avg_step=1.34s recent_step=1.15s eta=12:07:12 +2026-08-26 04:54:27 | INFO | train rank=0 | step=67340/100000 progress=67.3% loss=0.049133 lr=2.643563e-06 elapsed=24:58:57 avg_step=1.34s recent_step=1.37s eta=12:06:59 +2026-08-26 04:54:39 | INFO | train rank=0 | step=67350/100000 progress=67.3% loss=0.049133 lr=2.642105e-06 elapsed=24:59:09 avg_step=1.34s recent_step=1.21s eta=12:06:45 +2026-08-26 04:54:53 | INFO | train rank=0 | step=67360/100000 progress=67.4% loss=0.049698 lr=2.640647e-06 elapsed=24:59:23 avg_step=1.34s recent_step=1.39s eta=12:06:32 +2026-08-26 04:55:05 | INFO | train rank=0 | step=67370/100000 progress=67.4% loss=0.033905 lr=2.639189e-06 elapsed=24:59:34 avg_step=1.34s recent_step=1.19s eta=12:06:18 +2026-08-26 04:55:16 | INFO | train rank=0 | step=67380/100000 progress=67.4% loss=0.069962 lr=2.637732e-06 elapsed=24:59:46 avg_step=1.34s recent_step=1.18s eta=12:06:04 +2026-08-26 04:55:28 | INFO | train rank=0 | step=67390/100000 progress=67.4% loss=0.046860 lr=2.636275e-06 elapsed=24:59:58 avg_step=1.34s recent_step=1.16s eta=12:05:50 +2026-08-26 04:55:40 | INFO | train rank=0 | step=67400/100000 progress=67.4% loss=0.091797 lr=2.634818e-06 elapsed=25:00:09 avg_step=1.34s recent_step=1.16s eta=12:05:35 +2026-08-26 04:55:51 | INFO | train rank=0 | step=67410/100000 progress=67.4% loss=0.053848 lr=2.633361e-06 elapsed=25:00:21 avg_step=1.34s recent_step=1.17s eta=12:05:21 +2026-08-26 04:56:07 | INFO | train rank=0 | step=67420/100000 progress=67.4% loss=0.051552 lr=2.631905e-06 elapsed=25:00:37 avg_step=1.34s recent_step=1.62s eta=12:05:09 +2026-08-26 04:56:21 | INFO | train rank=0 | step=67430/100000 progress=67.4% loss=0.068390 lr=2.630449e-06 elapsed=25:00:51 avg_step=1.34s recent_step=1.31s eta=12:04:56 +2026-08-26 04:56:37 | INFO | train rank=0 | step=67440/100000 progress=67.4% loss=0.037827 lr=2.628993e-06 elapsed=25:01:07 avg_step=1.34s recent_step=1.61s eta=12:04:44 +2026-08-26 04:56:52 | INFO | train rank=0 | step=67450/100000 progress=67.5% loss=0.032036 lr=2.627537e-06 elapsed=25:01:22 avg_step=1.34s recent_step=1.52s eta=12:04:31 +2026-08-26 04:57:04 | INFO | train rank=0 | step=67460/100000 progress=67.5% loss=0.055634 lr=2.626082e-06 elapsed=25:01:34 avg_step=1.34s recent_step=1.23s eta=12:04:18 +2026-08-26 04:57:16 | INFO | train rank=0 | step=67470/100000 progress=67.5% loss=0.046280 lr=2.624627e-06 elapsed=25:01:46 avg_step=1.34s recent_step=1.18s eta=12:04:03 +2026-08-26 04:57:31 | INFO | train rank=0 | step=67480/100000 progress=67.5% loss=0.027153 lr=2.623172e-06 elapsed=25:02:00 avg_step=1.34s recent_step=1.45s eta=12:03:51 +2026-08-26 04:57:42 | INFO | train rank=0 | step=67490/100000 progress=67.5% loss=0.059479 lr=2.621717e-06 elapsed=25:02:12 avg_step=1.34s recent_step=1.18s eta=12:03:37 +2026-08-26 04:57:57 | INFO | train rank=0 | step=67500/100000 progress=67.5% loss=0.087891 lr=2.620263e-06 elapsed=25:02:27 avg_step=1.34s recent_step=1.45s eta=12:03:24 +2026-08-26 04:58:11 | INFO | train rank=0 | step=67510/100000 progress=67.5% loss=0.038284 lr=2.618809e-06 elapsed=25:02:41 avg_step=1.34s recent_step=1.40s eta=12:03:11 +2026-08-26 04:58:24 | INFO | train rank=0 | step=67520/100000 progress=67.5% loss=0.058746 lr=2.617355e-06 elapsed=25:02:54 avg_step=1.34s recent_step=1.30s eta=12:02:57 +2026-08-26 04:58:36 | INFO | train rank=0 | step=67530/100000 progress=67.5% loss=0.053650 lr=2.615902e-06 elapsed=25:03:06 avg_step=1.34s recent_step=1.26s eta=12:02:43 +2026-08-26 04:58:50 | INFO | train rank=0 | step=67540/100000 progress=67.5% loss=0.027161 lr=2.614448e-06 elapsed=25:03:20 avg_step=1.34s recent_step=1.34s eta=12:02:30 +2026-08-26 04:59:06 | INFO | train rank=0 | step=67550/100000 progress=67.5% loss=0.049866 lr=2.612995e-06 elapsed=25:03:36 avg_step=1.34s recent_step=1.59s eta=12:02:18 +2026-08-26 04:59:18 | INFO | train rank=0 | step=67560/100000 progress=67.6% loss=0.028427 lr=2.611543e-06 elapsed=25:03:48 avg_step=1.34s recent_step=1.20s eta=12:02:04 +2026-08-26 04:59:32 | INFO | train rank=0 | step=67570/100000 progress=67.6% loss=0.059723 lr=2.610090e-06 elapsed=25:04:02 avg_step=1.34s recent_step=1.39s eta=12:01:51 +2026-08-26 04:59:48 | INFO | train rank=0 | step=67580/100000 progress=67.6% loss=0.038513 lr=2.608638e-06 elapsed=25:04:17 avg_step=1.34s recent_step=1.59s eta=12:01:39 +2026-08-26 04:59:59 | INFO | train rank=0 | step=67590/100000 progress=67.6% loss=0.072144 lr=2.607186e-06 elapsed=25:04:29 avg_step=1.34s recent_step=1.12s eta=12:01:24 +2026-08-26 05:00:15 | INFO | train rank=0 | step=67600/100000 progress=67.6% loss=0.049240 lr=2.605734e-06 elapsed=25:04:45 avg_step=1.34s recent_step=1.60s eta=12:01:12 +2026-08-26 05:00:28 | INFO | train rank=0 | step=67610/100000 progress=67.6% loss=0.052628 lr=2.604283e-06 elapsed=25:04:57 avg_step=1.34s recent_step=1.28s eta=12:00:59 +2026-08-26 05:00:44 | INFO | train rank=0 | step=67620/100000 progress=67.6% loss=0.056541 lr=2.602832e-06 elapsed=25:05:13 avg_step=1.34s recent_step=1.60s eta=12:00:47 +2026-08-26 05:00:58 | INFO | train rank=0 | step=67630/100000 progress=67.6% loss=0.046432 lr=2.601381e-06 elapsed=25:05:28 avg_step=1.34s recent_step=1.49s eta=12:00:34 +2026-08-26 05:01:14 | INFO | train rank=0 | step=67640/100000 progress=67.6% loss=0.033310 lr=2.599930e-06 elapsed=25:05:44 avg_step=1.34s recent_step=1.55s eta=12:00:22 +2026-08-26 05:01:25 | INFO | train rank=0 | step=67650/100000 progress=67.7% loss=0.039185 lr=2.598480e-06 elapsed=25:05:55 avg_step=1.34s recent_step=1.15s eta=12:00:07 +2026-08-26 05:01:39 | INFO | train rank=0 | step=67660/100000 progress=67.7% loss=0.046249 lr=2.597030e-06 elapsed=25:06:09 avg_step=1.34s recent_step=1.39s eta=11:59:54 +2026-08-26 05:01:52 | INFO | train rank=0 | step=67670/100000 progress=67.7% loss=0.065994 lr=2.595580e-06 elapsed=25:06:21 avg_step=1.34s recent_step=1.21s eta=11:59:40 +2026-08-26 05:02:10 | INFO | train rank=0 | step=67680/100000 progress=67.7% loss=0.036270 lr=2.594130e-06 elapsed=25:06:39 avg_step=1.34s recent_step=1.80s eta=11:59:29 +2026-08-26 05:02:22 | INFO | train rank=0 | step=67690/100000 progress=67.7% loss=0.065430 lr=2.592681e-06 elapsed=25:06:52 avg_step=1.34s recent_step=1.24s eta=11:59:15 +2026-08-26 05:02:34 | INFO | train rank=0 | step=67700/100000 progress=67.7% loss=0.039261 lr=2.591232e-06 elapsed=25:07:04 avg_step=1.34s recent_step=1.17s eta=11:59:01 +2026-08-26 05:02:46 | INFO | train rank=0 | step=67710/100000 progress=67.7% loss=0.125000 lr=2.589783e-06 elapsed=25:07:15 avg_step=1.34s recent_step=1.19s eta=11:58:47 +2026-08-26 05:02:58 | INFO | train rank=0 | step=67720/100000 progress=67.7% loss=0.056366 lr=2.588334e-06 elapsed=25:07:28 avg_step=1.34s recent_step=1.24s eta=11:58:33 +2026-08-26 05:03:12 | INFO | train rank=0 | step=67730/100000 progress=67.7% loss=0.058746 lr=2.586886e-06 elapsed=25:07:42 avg_step=1.34s recent_step=1.46s eta=11:58:21 +2026-08-26 05:03:26 | INFO | train rank=0 | step=67740/100000 progress=67.7% loss=0.070984 lr=2.585438e-06 elapsed=25:07:56 avg_step=1.34s recent_step=1.34s eta=11:58:07 +2026-08-26 05:03:38 | INFO | train rank=0 | step=67750/100000 progress=67.8% loss=0.052307 lr=2.583990e-06 elapsed=25:08:08 avg_step=1.34s recent_step=1.20s eta=11:57:53 +2026-08-26 05:03:52 | INFO | train rank=0 | step=67760/100000 progress=67.8% loss=0.038437 lr=2.582543e-06 elapsed=25:08:22 avg_step=1.34s recent_step=1.40s eta=11:57:40 +2026-08-26 05:04:07 | INFO | train rank=0 | step=67770/100000 progress=67.8% loss=0.059998 lr=2.581095e-06 elapsed=25:08:37 avg_step=1.34s recent_step=1.53s eta=11:57:28 +2026-08-26 05:04:21 | INFO | train rank=0 | step=67780/100000 progress=67.8% loss=0.032318 lr=2.579648e-06 elapsed=25:08:51 avg_step=1.34s recent_step=1.36s eta=11:57:15 +2026-08-26 05:04:36 | INFO | train rank=0 | step=67790/100000 progress=67.8% loss=0.036957 lr=2.578202e-06 elapsed=25:09:06 avg_step=1.34s recent_step=1.54s eta=11:57:02 +2026-08-26 05:04:48 | INFO | train rank=0 | step=67800/100000 progress=67.8% loss=0.028076 lr=2.576755e-06 elapsed=25:09:18 avg_step=1.34s recent_step=1.16s eta=11:56:48 +2026-08-26 05:04:59 | INFO | train rank=0 | step=67810/100000 progress=67.8% loss=0.045090 lr=2.575309e-06 elapsed=25:09:29 avg_step=1.34s recent_step=1.15s eta=11:56:34 +2026-08-26 05:05:13 | INFO | train rank=0 | step=67820/100000 progress=67.8% loss=0.071503 lr=2.573863e-06 elapsed=25:09:43 avg_step=1.34s recent_step=1.38s eta=11:56:21 +2026-08-26 05:05:25 | INFO | train rank=0 | step=67830/100000 progress=67.8% loss=0.053131 lr=2.572418e-06 elapsed=25:09:55 avg_step=1.34s recent_step=1.17s eta=11:56:06 +2026-08-26 05:05:41 | INFO | train rank=0 | step=67840/100000 progress=67.8% loss=0.048279 lr=2.570972e-06 elapsed=25:10:11 avg_step=1.34s recent_step=1.65s eta=11:55:55 +2026-08-26 05:05:57 | INFO | train rank=0 | step=67850/100000 progress=67.8% loss=0.033783 lr=2.569527e-06 elapsed=25:10:27 avg_step=1.34s recent_step=1.61s eta=11:55:42 +2026-08-26 05:06:12 | INFO | train rank=0 | step=67860/100000 progress=67.9% loss=0.080627 lr=2.568082e-06 elapsed=25:10:41 avg_step=1.34s recent_step=1.42s eta=11:55:30 +2026-08-26 05:06:23 | INFO | train rank=0 | step=67870/100000 progress=67.9% loss=0.058823 lr=2.566638e-06 elapsed=25:10:53 avg_step=1.34s recent_step=1.15s eta=11:55:15 +2026-08-26 05:06:39 | INFO | train rank=0 | step=67880/100000 progress=67.9% loss=0.053680 lr=2.565193e-06 elapsed=25:11:09 avg_step=1.34s recent_step=1.59s eta=11:55:03 +2026-08-26 05:06:52 | INFO | train rank=0 | step=67890/100000 progress=67.9% loss=0.034790 lr=2.563749e-06 elapsed=25:11:22 avg_step=1.34s recent_step=1.30s eta=11:54:50 +2026-08-26 05:07:04 | INFO | train rank=0 | step=67900/100000 progress=67.9% loss=0.026436 lr=2.562306e-06 elapsed=25:11:34 avg_step=1.34s recent_step=1.18s eta=11:54:36 +2026-08-26 05:07:18 | INFO | train rank=0 | step=67910/100000 progress=67.9% loss=0.062775 lr=2.560862e-06 elapsed=25:11:48 avg_step=1.34s recent_step=1.41s eta=11:54:23 +2026-08-26 05:07:30 | INFO | train rank=0 | step=67920/100000 progress=67.9% loss=0.049484 lr=2.559419e-06 elapsed=25:12:00 avg_step=1.34s recent_step=1.17s eta=11:54:08 +2026-08-26 05:07:44 | INFO | train rank=0 | step=67930/100000 progress=67.9% loss=0.040909 lr=2.557976e-06 elapsed=25:12:14 avg_step=1.34s recent_step=1.44s eta=11:53:56 +2026-08-26 05:07:59 | INFO | train rank=0 | step=67940/100000 progress=67.9% loss=0.029442 lr=2.556533e-06 elapsed=25:12:29 avg_step=1.34s recent_step=1.49s eta=11:53:43 +2026-08-26 05:08:13 | INFO | train rank=0 | step=67950/100000 progress=68.0% loss=0.060028 lr=2.555091e-06 elapsed=25:12:43 avg_step=1.34s recent_step=1.42s eta=11:53:30 +2026-08-26 05:08:27 | INFO | train rank=0 | step=67960/100000 progress=68.0% loss=0.047714 lr=2.553649e-06 elapsed=25:12:57 avg_step=1.34s recent_step=1.37s eta=11:53:17 +2026-08-26 05:08:38 | INFO | train rank=0 | step=67970/100000 progress=68.0% loss=0.043777 lr=2.552207e-06 elapsed=25:13:08 avg_step=1.34s recent_step=1.14s eta=11:53:03 +2026-08-26 05:08:50 | INFO | train rank=0 | step=67980/100000 progress=68.0% loss=0.041168 lr=2.550765e-06 elapsed=25:13:19 avg_step=1.34s recent_step=1.12s eta=11:52:48 +2026-08-26 05:09:04 | INFO | train rank=0 | step=67990/100000 progress=68.0% loss=0.044342 lr=2.549324e-06 elapsed=25:13:33 avg_step=1.34s recent_step=1.41s eta=11:52:35 +2026-08-26 05:09:19 | INFO | train rank=0 | step=68000/100000 progress=68.0% loss=0.073181 lr=2.547882e-06 elapsed=25:13:49 avg_step=1.34s recent_step=1.55s eta=11:52:23 +2026-08-26 05:09:30 | INFO | train rank=0 | step=68010/100000 progress=68.0% loss=0.046356 lr=2.546442e-06 elapsed=25:14:00 avg_step=1.34s recent_step=1.14s eta=11:52:08 +2026-08-26 05:09:42 | INFO | train rank=0 | step=68020/100000 progress=68.0% loss=0.048584 lr=2.545001e-06 elapsed=25:14:12 avg_step=1.34s recent_step=1.20s eta=11:51:54 +2026-08-26 05:09:59 | INFO | train rank=0 | step=68030/100000 progress=68.0% loss=0.033340 lr=2.543561e-06 elapsed=25:14:29 avg_step=1.34s recent_step=1.63s eta=11:51:43 +2026-08-26 05:10:10 | INFO | train rank=0 | step=68040/100000 progress=68.0% loss=0.048492 lr=2.542121e-06 elapsed=25:14:40 avg_step=1.34s recent_step=1.13s eta=11:51:28 +2026-08-26 05:10:22 | INFO | train rank=0 | step=68050/100000 progress=68.0% loss=0.060165 lr=2.540681e-06 elapsed=25:14:52 avg_step=1.34s recent_step=1.22s eta=11:51:14 +2026-08-26 05:10:36 | INFO | train rank=0 | step=68060/100000 progress=68.1% loss=0.049759 lr=2.539241e-06 elapsed=25:15:06 avg_step=1.34s recent_step=1.40s eta=11:51:01 +2026-08-26 05:10:52 | INFO | train rank=0 | step=68070/100000 progress=68.1% loss=0.065613 lr=2.537802e-06 elapsed=25:15:22 avg_step=1.34s recent_step=1.61s eta=11:50:49 +2026-08-26 05:11:05 | INFO | train rank=0 | step=68080/100000 progress=68.1% loss=0.032551 lr=2.536363e-06 elapsed=25:15:35 avg_step=1.34s recent_step=1.26s eta=11:50:35 +2026-08-26 05:11:16 | INFO | train rank=0 | step=68090/100000 progress=68.1% loss=0.050354 lr=2.534925e-06 elapsed=25:15:46 avg_step=1.34s recent_step=1.14s eta=11:50:21 +2026-08-26 05:11:30 | INFO | train rank=0 | step=68100/100000 progress=68.1% loss=0.040436 lr=2.533486e-06 elapsed=25:16:00 avg_step=1.34s recent_step=1.41s eta=11:50:08 +2026-08-26 05:11:42 | INFO | train rank=0 | step=68110/100000 progress=68.1% loss=0.051262 lr=2.532048e-06 elapsed=25:16:12 avg_step=1.34s recent_step=1.19s eta=11:49:54 +2026-08-26 05:11:55 | INFO | train rank=0 | step=68120/100000 progress=68.1% loss=0.056396 lr=2.530610e-06 elapsed=25:16:25 avg_step=1.34s recent_step=1.28s eta=11:49:40 +2026-08-26 05:12:08 | INFO | train rank=0 | step=68130/100000 progress=68.1% loss=0.043991 lr=2.529173e-06 elapsed=25:16:37 avg_step=1.34s recent_step=1.25s eta=11:49:27 +2026-08-26 05:12:22 | INFO | train rank=0 | step=68140/100000 progress=68.1% loss=0.050491 lr=2.527735e-06 elapsed=25:16:52 avg_step=1.34s recent_step=1.43s eta=11:49:14 +2026-08-26 05:12:37 | INFO | train rank=0 | step=68150/100000 progress=68.2% loss=0.048630 lr=2.526298e-06 elapsed=25:17:07 avg_step=1.34s recent_step=1.52s eta=11:49:01 +2026-08-26 05:12:52 | INFO | train rank=0 | step=68160/100000 progress=68.2% loss=0.037125 lr=2.524861e-06 elapsed=25:17:22 avg_step=1.34s recent_step=1.49s eta=11:48:49 +2026-08-26 05:13:06 | INFO | train rank=0 | step=68170/100000 progress=68.2% loss=0.043961 lr=2.523425e-06 elapsed=25:17:36 avg_step=1.34s recent_step=1.44s eta=11:48:36 +2026-08-26 05:13:21 | INFO | train rank=0 | step=68180/100000 progress=68.2% loss=0.043091 lr=2.521989e-06 elapsed=25:17:51 avg_step=1.34s recent_step=1.46s eta=11:48:23 +2026-08-26 05:13:33 | INFO | train rank=0 | step=68190/100000 progress=68.2% loss=0.043396 lr=2.520553e-06 elapsed=25:18:03 avg_step=1.34s recent_step=1.21s eta=11:48:09 +2026-08-26 05:13:44 | INFO | train rank=0 | step=68200/100000 progress=68.2% loss=0.030373 lr=2.519117e-06 elapsed=25:18:14 avg_step=1.34s recent_step=1.12s eta=11:47:55 +2026-08-26 05:13:56 | INFO | train rank=0 | step=68210/100000 progress=68.2% loss=0.084503 lr=2.517681e-06 elapsed=25:18:26 avg_step=1.34s recent_step=1.19s eta=11:47:41 +2026-08-26 05:14:10 | INFO | train rank=0 | step=68220/100000 progress=68.2% loss=0.053436 lr=2.516246e-06 elapsed=25:18:40 avg_step=1.34s recent_step=1.36s eta=11:47:27 +2026-08-26 05:14:22 | INFO | train rank=0 | step=68230/100000 progress=68.2% loss=0.032188 lr=2.514811e-06 elapsed=25:18:52 avg_step=1.34s recent_step=1.22s eta=11:47:14 +2026-08-26 05:14:38 | INFO | train rank=0 | step=68240/100000 progress=68.2% loss=0.040497 lr=2.513377e-06 elapsed=25:19:08 avg_step=1.34s recent_step=1.65s eta=11:47:02 +2026-08-26 05:14:51 | INFO | train rank=0 | step=68250/100000 progress=68.2% loss=0.051682 lr=2.511942e-06 elapsed=25:19:21 avg_step=1.34s recent_step=1.31s eta=11:46:48 +2026-08-26 05:15:03 | INFO | train rank=0 | step=68260/100000 progress=68.3% loss=0.056778 lr=2.510508e-06 elapsed=25:19:33 avg_step=1.34s recent_step=1.15s eta=11:46:34 +2026-08-26 05:15:17 | INFO | train rank=0 | step=68270/100000 progress=68.3% loss=0.038696 lr=2.509074e-06 elapsed=25:19:47 avg_step=1.34s recent_step=1.44s eta=11:46:21 +2026-08-26 05:15:32 | INFO | train rank=0 | step=68280/100000 progress=68.3% loss=0.052322 lr=2.507641e-06 elapsed=25:20:02 avg_step=1.34s recent_step=1.49s eta=11:46:08 +2026-08-26 05:15:44 | INFO | train rank=0 | step=68290/100000 progress=68.3% loss=0.029877 lr=2.506208e-06 elapsed=25:20:14 avg_step=1.34s recent_step=1.15s eta=11:45:54 +2026-08-26 05:15:58 | INFO | train rank=0 | step=68300/100000 progress=68.3% loss=0.049179 lr=2.504775e-06 elapsed=25:20:28 avg_step=1.34s recent_step=1.37s eta=11:45:41 +2026-08-26 05:16:09 | INFO | train rank=0 | step=68310/100000 progress=68.3% loss=0.071838 lr=2.503342e-06 elapsed=25:20:39 avg_step=1.34s recent_step=1.14s eta=11:45:27 +2026-08-26 05:16:27 | INFO | train rank=0 | step=68320/100000 progress=68.3% loss=0.063904 lr=2.501910e-06 elapsed=25:20:57 avg_step=1.34s recent_step=1.80s eta=11:45:16 +2026-08-26 05:16:39 | INFO | train rank=0 | step=68330/100000 progress=68.3% loss=0.051865 lr=2.500477e-06 elapsed=25:21:09 avg_step=1.34s recent_step=1.22s eta=11:45:02 +2026-08-26 05:16:51 | INFO | train rank=0 | step=68340/100000 progress=68.3% loss=0.040039 lr=2.499045e-06 elapsed=25:21:21 avg_step=1.34s recent_step=1.14s eta=11:44:47 +2026-08-26 05:17:03 | INFO | train rank=0 | step=68350/100000 progress=68.3% loss=0.036636 lr=2.497614e-06 elapsed=25:21:33 avg_step=1.34s recent_step=1.20s eta=11:44:33 +2026-08-26 05:17:16 | INFO | train rank=0 | step=68360/100000 progress=68.4% loss=0.045990 lr=2.496182e-06 elapsed=25:21:46 avg_step=1.34s recent_step=1.36s eta=11:44:20 +2026-08-26 05:17:30 | INFO | train rank=0 | step=68370/100000 progress=68.4% loss=0.045410 lr=2.494751e-06 elapsed=25:22:00 avg_step=1.34s recent_step=1.41s eta=11:44:07 +2026-08-26 05:17:42 | INFO | train rank=0 | step=68380/100000 progress=68.4% loss=0.036758 lr=2.493321e-06 elapsed=25:22:12 avg_step=1.34s recent_step=1.19s eta=11:43:53 +2026-08-26 05:17:54 | INFO | train rank=0 | step=68390/100000 progress=68.4% loss=0.075043 lr=2.491890e-06 elapsed=25:22:24 avg_step=1.34s recent_step=1.16s eta=11:43:39 +2026-08-26 05:18:09 | INFO | train rank=0 | step=68400/100000 progress=68.4% loss=0.046097 lr=2.490460e-06 elapsed=25:22:39 avg_step=1.34s recent_step=1.53s eta=11:43:27 +2026-08-26 05:18:24 | INFO | train rank=0 | step=68410/100000 progress=68.4% loss=0.041992 lr=2.489030e-06 elapsed=25:22:54 avg_step=1.34s recent_step=1.46s eta=11:43:14 +2026-08-26 05:18:35 | INFO | train rank=0 | step=68420/100000 progress=68.4% loss=0.050201 lr=2.487600e-06 elapsed=25:23:05 avg_step=1.34s recent_step=1.13s eta=11:43:00 +2026-08-26 05:18:48 | INFO | train rank=0 | step=68430/100000 progress=68.4% loss=0.066864 lr=2.486171e-06 elapsed=25:23:18 avg_step=1.34s recent_step=1.33s eta=11:42:46 +2026-08-26 05:19:00 | INFO | train rank=0 | step=68440/100000 progress=68.4% loss=0.044342 lr=2.484741e-06 elapsed=25:23:30 avg_step=1.34s recent_step=1.19s eta=11:42:32 +2026-08-26 05:19:12 | INFO | train rank=0 | step=68450/100000 progress=68.5% loss=0.043640 lr=2.483313e-06 elapsed=25:23:42 avg_step=1.34s recent_step=1.18s eta=11:42:18 +2026-08-26 05:19:27 | INFO | train rank=0 | step=68460/100000 progress=68.5% loss=0.045509 lr=2.481884e-06 elapsed=25:23:57 avg_step=1.34s recent_step=1.55s eta=11:42:06 +2026-08-26 05:19:41 | INFO | train rank=0 | step=68470/100000 progress=68.5% loss=0.029472 lr=2.480456e-06 elapsed=25:24:11 avg_step=1.34s recent_step=1.39s eta=11:41:52 +2026-08-26 05:19:57 | INFO | train rank=0 | step=68480/100000 progress=68.5% loss=0.055603 lr=2.479028e-06 elapsed=25:24:27 avg_step=1.34s recent_step=1.61s eta=11:41:40 +2026-08-26 05:20:09 | INFO | train rank=0 | step=68490/100000 progress=68.5% loss=0.054642 lr=2.477600e-06 elapsed=25:24:39 avg_step=1.34s recent_step=1.19s eta=11:41:26 +2026-08-26 05:20:22 | INFO | train rank=0 | step=68500/100000 progress=68.5% loss=0.042007 lr=2.476172e-06 elapsed=25:24:52 avg_step=1.34s recent_step=1.29s eta=11:41:13 +2026-08-26 05:20:35 | INFO | train rank=0 | step=68510/100000 progress=68.5% loss=0.063171 lr=2.474745e-06 elapsed=25:25:05 avg_step=1.34s recent_step=1.23s eta=11:40:59 +2026-08-26 05:20:52 | INFO | train rank=0 | step=68520/100000 progress=68.5% loss=0.057693 lr=2.473318e-06 elapsed=25:25:22 avg_step=1.34s recent_step=1.78s eta=11:40:48 +2026-08-26 05:21:06 | INFO | train rank=0 | step=68530/100000 progress=68.5% loss=0.046326 lr=2.471891e-06 elapsed=25:25:36 avg_step=1.34s recent_step=1.38s eta=11:40:35 +2026-08-26 05:21:20 | INFO | train rank=0 | step=68540/100000 progress=68.5% loss=0.039108 lr=2.470465e-06 elapsed=25:25:50 avg_step=1.34s recent_step=1.42s eta=11:40:22 +2026-08-26 05:21:35 | INFO | train rank=0 | step=68550/100000 progress=68.5% loss=0.091614 lr=2.469039e-06 elapsed=25:26:05 avg_step=1.34s recent_step=1.44s eta=11:40:09 +2026-08-26 05:21:47 | INFO | train rank=0 | step=68560/100000 progress=68.6% loss=0.068787 lr=2.467613e-06 elapsed=25:26:17 avg_step=1.34s recent_step=1.21s eta=11:39:55 +2026-08-26 05:22:01 | INFO | train rank=0 | step=68570/100000 progress=68.6% loss=0.042404 lr=2.466188e-06 elapsed=25:26:31 avg_step=1.34s recent_step=1.39s eta=11:39:42 +2026-08-26 05:22:12 | INFO | train rank=0 | step=68580/100000 progress=68.6% loss=0.042313 lr=2.464762e-06 elapsed=25:26:42 avg_step=1.34s recent_step=1.15s eta=11:39:27 +2026-08-26 05:22:24 | INFO | train rank=0 | step=68590/100000 progress=68.6% loss=0.032616 lr=2.463337e-06 elapsed=25:26:54 avg_step=1.34s recent_step=1.20s eta=11:39:13 +2026-08-26 05:22:44 | INFO | train rank=0 | step=68600/100000 progress=68.6% loss=0.060226 lr=2.461912e-06 elapsed=25:27:14 avg_step=1.34s recent_step=1.94s eta=11:39:03 +2026-08-26 05:22:55 | INFO | train rank=0 | step=68610/100000 progress=68.6% loss=0.035393 lr=2.460488e-06 elapsed=25:27:25 avg_step=1.34s recent_step=1.10s eta=11:38:48 +2026-08-26 05:23:09 | INFO | train rank=0 | step=68620/100000 progress=68.6% loss=0.040619 lr=2.459064e-06 elapsed=25:27:39 avg_step=1.34s recent_step=1.41s eta=11:38:35 +2026-08-26 05:23:20 | INFO | train rank=0 | step=68630/100000 progress=68.6% loss=0.043953 lr=2.457640e-06 elapsed=25:27:50 avg_step=1.34s recent_step=1.15s eta=11:38:21 +2026-08-26 05:23:37 | INFO | train rank=0 | step=68640/100000 progress=68.6% loss=0.047577 lr=2.456216e-06 elapsed=25:28:07 avg_step=1.34s recent_step=1.67s eta=11:38:09 +2026-08-26 05:23:49 | INFO | train rank=0 | step=68650/100000 progress=68.7% loss=0.042366 lr=2.454793e-06 elapsed=25:28:19 avg_step=1.34s recent_step=1.17s eta=11:37:55 +2026-08-26 05:24:00 | INFO | train rank=0 | step=68660/100000 progress=68.7% loss=0.062668 lr=2.453370e-06 elapsed=25:28:30 avg_step=1.34s recent_step=1.15s eta=11:37:41 +2026-08-26 05:24:12 | INFO | train rank=0 | step=68670/100000 progress=68.7% loss=0.043930 lr=2.451947e-06 elapsed=25:28:42 avg_step=1.34s recent_step=1.19s eta=11:37:27 +2026-08-26 05:24:25 | INFO | train rank=0 | step=68680/100000 progress=68.7% loss=0.069092 lr=2.450525e-06 elapsed=25:28:55 avg_step=1.34s recent_step=1.32s eta=11:37:14 +2026-08-26 05:24:41 | INFO | train rank=0 | step=68690/100000 progress=68.7% loss=0.041245 lr=2.449102e-06 elapsed=25:29:11 avg_step=1.34s recent_step=1.57s eta=11:37:01 +2026-08-26 05:24:53 | INFO | train rank=0 | step=68700/100000 progress=68.7% loss=0.053543 lr=2.447680e-06 elapsed=25:29:23 avg_step=1.34s recent_step=1.19s eta=11:36:47 +2026-08-26 05:25:04 | INFO | train rank=0 | step=68710/100000 progress=68.7% loss=0.062164 lr=2.446259e-06 elapsed=25:29:34 avg_step=1.34s recent_step=1.14s eta=11:36:33 +2026-08-26 05:25:21 | INFO | train rank=0 | step=68720/100000 progress=68.7% loss=0.060089 lr=2.444837e-06 elapsed=25:29:51 avg_step=1.34s recent_step=1.71s eta=11:36:21 +2026-08-26 05:25:33 | INFO | train rank=0 | step=68730/100000 progress=68.7% loss=0.043594 lr=2.443416e-06 elapsed=25:30:03 avg_step=1.34s recent_step=1.20s eta=11:36:07 +2026-08-26 05:25:49 | INFO | train rank=0 | step=68740/100000 progress=68.7% loss=0.051025 lr=2.441995e-06 elapsed=25:30:19 avg_step=1.34s recent_step=1.57s eta=11:35:55 +2026-08-26 05:26:01 | INFO | train rank=0 | step=68750/100000 progress=68.8% loss=0.042664 lr=2.440575e-06 elapsed=25:30:31 avg_step=1.34s recent_step=1.23s eta=11:35:41 +2026-08-26 05:26:13 | INFO | train rank=0 | step=68760/100000 progress=68.8% loss=0.046959 lr=2.439154e-06 elapsed=25:30:43 avg_step=1.34s recent_step=1.16s eta=11:35:27 +2026-08-26 05:26:26 | INFO | train rank=0 | step=68770/100000 progress=68.8% loss=0.054825 lr=2.437734e-06 elapsed=25:30:56 avg_step=1.34s recent_step=1.34s eta=11:35:14 +2026-08-26 05:26:40 | INFO | train rank=0 | step=68780/100000 progress=68.8% loss=0.034058 lr=2.436315e-06 elapsed=25:31:10 avg_step=1.34s recent_step=1.36s eta=11:35:00 +2026-08-26 05:26:54 | INFO | train rank=0 | step=68790/100000 progress=68.8% loss=0.032860 lr=2.434895e-06 elapsed=25:31:24 avg_step=1.34s recent_step=1.38s eta=11:34:47 +2026-08-26 05:27:06 | INFO | train rank=0 | step=68800/100000 progress=68.8% loss=0.061890 lr=2.433476e-06 elapsed=25:31:36 avg_step=1.34s recent_step=1.19s eta=11:34:33 +2026-08-26 05:27:20 | INFO | train rank=0 | step=68810/100000 progress=68.8% loss=0.032562 lr=2.432057e-06 elapsed=25:31:50 avg_step=1.34s recent_step=1.43s eta=11:34:20 +2026-08-26 05:27:32 | INFO | train rank=0 | step=68820/100000 progress=68.8% loss=0.031296 lr=2.430639e-06 elapsed=25:32:02 avg_step=1.34s recent_step=1.18s eta=11:34:06 +2026-08-26 05:27:44 | INFO | train rank=0 | step=68830/100000 progress=68.8% loss=0.062973 lr=2.429220e-06 elapsed=25:32:13 avg_step=1.34s recent_step=1.17s eta=11:33:52 +2026-08-26 05:27:55 | INFO | train rank=0 | step=68840/100000 progress=68.8% loss=0.074066 lr=2.427802e-06 elapsed=25:32:25 avg_step=1.34s recent_step=1.18s eta=11:33:38 +2026-08-26 05:28:10 | INFO | train rank=0 | step=68850/100000 progress=68.8% loss=0.038498 lr=2.426385e-06 elapsed=25:32:40 avg_step=1.34s recent_step=1.45s eta=11:33:25 +2026-08-26 05:28:25 | INFO | train rank=0 | step=68860/100000 progress=68.9% loss=0.074188 lr=2.424967e-06 elapsed=25:32:55 avg_step=1.34s recent_step=1.48s eta=11:33:13 +2026-08-26 05:28:35 | INFO | train rank=0 | step=68870/100000 progress=68.9% loss=0.049454 lr=2.423550e-06 elapsed=25:33:05 avg_step=1.34s recent_step=1.08s eta=11:32:58 +2026-08-26 05:28:49 | INFO | train rank=0 | step=68880/100000 progress=68.9% loss=0.033188 lr=2.422133e-06 elapsed=25:33:19 avg_step=1.34s recent_step=1.37s eta=11:32:45 +2026-08-26 05:29:01 | INFO | train rank=0 | step=68890/100000 progress=68.9% loss=0.053421 lr=2.420716e-06 elapsed=25:33:31 avg_step=1.34s recent_step=1.22s eta=11:32:31 +2026-08-26 05:29:13 | INFO | train rank=0 | step=68900/100000 progress=68.9% loss=0.035736 lr=2.419300e-06 elapsed=25:33:43 avg_step=1.34s recent_step=1.18s eta=11:32:17 +2026-08-26 05:29:28 | INFO | train rank=0 | step=68910/100000 progress=68.9% loss=0.039520 lr=2.417884e-06 elapsed=25:33:58 avg_step=1.34s recent_step=1.53s eta=11:32:04 +2026-08-26 05:29:41 | INFO | train rank=0 | step=68920/100000 progress=68.9% loss=0.036041 lr=2.416468e-06 elapsed=25:34:11 avg_step=1.34s recent_step=1.24s eta=11:31:51 +2026-08-26 05:29:54 | INFO | train rank=0 | step=68930/100000 progress=68.9% loss=0.055786 lr=2.415053e-06 elapsed=25:34:24 avg_step=1.34s recent_step=1.34s eta=11:31:37 +2026-08-26 05:30:10 | INFO | train rank=0 | step=68940/100000 progress=68.9% loss=0.032944 lr=2.413637e-06 elapsed=25:34:40 avg_step=1.34s recent_step=1.54s eta=11:31:25 +2026-08-26 05:30:22 | INFO | train rank=0 | step=68950/100000 progress=69.0% loss=0.048615 lr=2.412223e-06 elapsed=25:34:52 avg_step=1.34s recent_step=1.26s eta=11:31:11 +2026-08-26 05:30:34 | INFO | train rank=0 | step=68960/100000 progress=69.0% loss=0.051239 lr=2.410808e-06 elapsed=25:35:04 avg_step=1.34s recent_step=1.18s eta=11:30:57 +2026-08-26 05:30:51 | INFO | train rank=0 | step=68970/100000 progress=69.0% loss=0.040939 lr=2.409394e-06 elapsed=25:35:21 avg_step=1.34s recent_step=1.70s eta=11:30:45 +2026-08-26 05:31:03 | INFO | train rank=0 | step=68980/100000 progress=69.0% loss=0.040260 lr=2.407979e-06 elapsed=25:35:32 avg_step=1.34s recent_step=1.15s eta=11:30:31 +2026-08-26 05:31:16 | INFO | train rank=0 | step=68990/100000 progress=69.0% loss=0.044968 lr=2.406566e-06 elapsed=25:35:46 avg_step=1.34s recent_step=1.33s eta=11:30:18 +2026-08-26 05:31:34 | INFO | train rank=0 | step=69000/100000 progress=69.0% loss=0.045898 lr=2.405152e-06 elapsed=25:36:03 avg_step=1.34s recent_step=1.77s eta=11:30:06 +2026-08-26 05:31:45 | INFO | train rank=0 | step=69010/100000 progress=69.0% loss=0.060120 lr=2.403739e-06 elapsed=25:36:15 avg_step=1.34s recent_step=1.17s eta=11:29:52 +2026-08-26 05:31:59 | INFO | train rank=0 | step=69020/100000 progress=69.0% loss=0.047760 lr=2.402326e-06 elapsed=25:36:29 avg_step=1.34s recent_step=1.34s eta=11:29:39 +2026-08-26 05:32:10 | INFO | train rank=0 | step=69030/100000 progress=69.0% loss=0.058960 lr=2.400913e-06 elapsed=25:36:40 avg_step=1.34s recent_step=1.16s eta=11:29:25 +2026-08-26 05:32:23 | INFO | train rank=0 | step=69040/100000 progress=69.0% loss=0.043320 lr=2.399501e-06 elapsed=25:36:52 avg_step=1.34s recent_step=1.22s eta=11:29:11 +2026-08-26 05:32:37 | INFO | train rank=0 | step=69050/100000 progress=69.0% loss=0.082123 lr=2.398089e-06 elapsed=25:37:07 avg_step=1.34s recent_step=1.43s eta=11:28:58 +2026-08-26 05:32:49 | INFO | train rank=0 | step=69060/100000 progress=69.1% loss=0.039001 lr=2.396677e-06 elapsed=25:37:19 avg_step=1.34s recent_step=1.24s eta=11:28:44 +2026-08-26 05:33:04 | INFO | train rank=0 | step=69070/100000 progress=69.1% loss=0.046585 lr=2.395265e-06 elapsed=25:37:33 avg_step=1.34s recent_step=1.44s eta=11:28:31 +2026-08-26 05:33:20 | INFO | train rank=0 | step=69080/100000 progress=69.1% loss=0.027008 lr=2.393854e-06 elapsed=25:37:50 avg_step=1.34s recent_step=1.66s eta=11:28:20 +2026-08-26 05:33:33 | INFO | train rank=0 | step=69090/100000 progress=69.1% loss=0.052200 lr=2.392443e-06 elapsed=25:38:02 avg_step=1.34s recent_step=1.24s eta=11:28:06 +2026-08-26 05:33:47 | INFO | train rank=0 | step=69100/100000 progress=69.1% loss=0.030075 lr=2.391033e-06 elapsed=25:38:17 avg_step=1.34s recent_step=1.46s eta=11:27:53 +2026-08-26 05:34:03 | INFO | train rank=0 | step=69110/100000 progress=69.1% loss=0.040031 lr=2.389622e-06 elapsed=25:38:33 avg_step=1.34s recent_step=1.58s eta=11:27:41 +2026-08-26 05:34:19 | INFO | train rank=0 | step=69120/100000 progress=69.1% loss=0.051346 lr=2.388212e-06 elapsed=25:38:49 avg_step=1.34s recent_step=1.59s eta=11:27:28 +2026-08-26 05:34:31 | INFO | train rank=0 | step=69130/100000 progress=69.1% loss=0.047867 lr=2.386802e-06 elapsed=25:39:01 avg_step=1.34s recent_step=1.18s eta=11:27:14 +2026-08-26 05:34:42 | INFO | train rank=0 | step=69140/100000 progress=69.1% loss=0.041374 lr=2.385393e-06 elapsed=25:39:12 avg_step=1.34s recent_step=1.13s eta=11:27:00 +2026-08-26 05:34:54 | INFO | train rank=0 | step=69150/100000 progress=69.2% loss=0.071106 lr=2.383984e-06 elapsed=25:39:24 avg_step=1.34s recent_step=1.22s eta=11:26:46 +2026-08-26 05:35:07 | INFO | train rank=0 | step=69160/100000 progress=69.2% loss=0.039200 lr=2.382575e-06 elapsed=25:39:37 avg_step=1.34s recent_step=1.27s eta=11:26:33 +2026-08-26 05:35:22 | INFO | train rank=0 | step=69170/100000 progress=69.2% loss=0.041145 lr=2.381166e-06 elapsed=25:39:52 avg_step=1.34s recent_step=1.50s eta=11:26:20 +2026-08-26 05:35:35 | INFO | train rank=0 | step=69180/100000 progress=69.2% loss=0.050613 lr=2.379758e-06 elapsed=25:40:05 avg_step=1.34s recent_step=1.30s eta=11:26:06 +2026-08-26 05:35:49 | INFO | train rank=0 | step=69190/100000 progress=69.2% loss=0.049301 lr=2.378349e-06 elapsed=25:40:19 avg_step=1.34s recent_step=1.42s eta=11:25:53 +2026-08-26 05:36:03 | INFO | train rank=0 | step=69200/100000 progress=69.2% loss=0.048233 lr=2.376942e-06 elapsed=25:40:33 avg_step=1.34s recent_step=1.42s eta=11:25:40 +2026-08-26 05:36:20 | INFO | train rank=0 | step=69210/100000 progress=69.2% loss=0.061127 lr=2.375534e-06 elapsed=25:40:50 avg_step=1.34s recent_step=1.64s eta=11:25:28 +2026-08-26 05:36:31 | INFO | train rank=0 | step=69220/100000 progress=69.2% loss=0.027260 lr=2.374127e-06 elapsed=25:41:01 avg_step=1.34s recent_step=1.14s eta=11:25:14 +2026-08-26 05:36:45 | INFO | train rank=0 | step=69230/100000 progress=69.2% loss=0.045776 lr=2.372720e-06 elapsed=25:41:15 avg_step=1.34s recent_step=1.37s eta=11:25:01 +2026-08-26 05:36:57 | INFO | train rank=0 | step=69240/100000 progress=69.2% loss=0.040382 lr=2.371313e-06 elapsed=25:41:27 avg_step=1.34s recent_step=1.26s eta=11:24:47 +2026-08-26 05:37:11 | INFO | train rank=0 | step=69250/100000 progress=69.2% loss=0.050201 lr=2.369907e-06 elapsed=25:41:41 avg_step=1.34s recent_step=1.38s eta=11:24:34 +2026-08-26 05:37:25 | INFO | train rank=0 | step=69260/100000 progress=69.3% loss=0.048035 lr=2.368501e-06 elapsed=25:41:54 avg_step=1.34s recent_step=1.34s eta=11:24:21 +2026-08-26 05:37:39 | INFO | train rank=0 | step=69270/100000 progress=69.3% loss=0.062805 lr=2.367095e-06 elapsed=25:42:09 avg_step=1.34s recent_step=1.46s eta=11:24:08 +2026-08-26 05:37:54 | INFO | train rank=0 | step=69280/100000 progress=69.3% loss=0.043869 lr=2.365689e-06 elapsed=25:42:24 avg_step=1.34s recent_step=1.48s eta=11:23:55 +2026-08-26 05:38:05 | INFO | train rank=0 | step=69290/100000 progress=69.3% loss=0.038864 lr=2.364284e-06 elapsed=25:42:35 avg_step=1.34s recent_step=1.14s eta=11:23:41 +2026-08-26 05:38:23 | INFO | train rank=0 | step=69300/100000 progress=69.3% loss=0.053223 lr=2.362879e-06 elapsed=25:42:53 avg_step=1.34s recent_step=1.75s eta=11:23:30 +2026-08-26 05:38:35 | INFO | train rank=0 | step=69310/100000 progress=69.3% loss=0.033638 lr=2.361475e-06 elapsed=25:43:05 avg_step=1.34s recent_step=1.22s eta=11:23:16 +2026-08-26 05:38:46 | INFO | train rank=0 | step=69320/100000 progress=69.3% loss=0.029350 lr=2.360070e-06 elapsed=25:43:16 avg_step=1.34s recent_step=1.10s eta=11:23:01 +2026-08-26 05:39:01 | INFO | train rank=0 | step=69330/100000 progress=69.3% loss=0.028183 lr=2.358666e-06 elapsed=25:43:31 avg_step=1.34s recent_step=1.50s eta=11:22:49 +2026-08-26 05:39:15 | INFO | train rank=0 | step=69340/100000 progress=69.3% loss=0.036911 lr=2.357262e-06 elapsed=25:43:44 avg_step=1.34s recent_step=1.34s eta=11:22:35 +2026-08-26 05:39:26 | INFO | train rank=0 | step=69350/100000 progress=69.3% loss=0.047554 lr=2.355859e-06 elapsed=25:43:56 avg_step=1.34s recent_step=1.13s eta=11:22:21 +2026-08-26 05:39:42 | INFO | train rank=0 | step=69360/100000 progress=69.4% loss=0.046829 lr=2.354456e-06 elapsed=25:44:12 avg_step=1.34s recent_step=1.58s eta=11:22:09 +2026-08-26 05:39:53 | INFO | train rank=0 | step=69370/100000 progress=69.4% loss=0.041061 lr=2.353053e-06 elapsed=25:44:23 avg_step=1.34s recent_step=1.16s eta=11:21:55 +2026-08-26 05:40:04 | INFO | train rank=0 | step=69380/100000 progress=69.4% loss=0.027565 lr=2.351650e-06 elapsed=25:44:34 avg_step=1.34s recent_step=1.11s eta=11:21:40 +2026-08-26 05:40:22 | INFO | train rank=0 | step=69390/100000 progress=69.4% loss=0.070587 lr=2.350248e-06 elapsed=25:44:52 avg_step=1.34s recent_step=1.81s eta=11:21:29 +2026-08-26 05:40:35 | INFO | train rank=0 | step=69400/100000 progress=69.4% loss=0.058243 lr=2.348846e-06 elapsed=25:45:04 avg_step=1.34s recent_step=1.21s eta=11:21:15 +2026-08-26 05:40:51 | INFO | train rank=0 | step=69410/100000 progress=69.4% loss=0.046326 lr=2.347444e-06 elapsed=25:45:21 avg_step=1.34s recent_step=1.63s eta=11:21:03 +2026-08-26 05:41:02 | INFO | train rank=0 | step=69420/100000 progress=69.4% loss=0.056473 lr=2.346043e-06 elapsed=25:45:32 avg_step=1.34s recent_step=1.16s eta=11:20:49 +2026-08-26 05:41:14 | INFO | train rank=0 | step=69430/100000 progress=69.4% loss=0.079071 lr=2.344641e-06 elapsed=25:45:44 avg_step=1.34s recent_step=1.17s eta=11:20:35 +2026-08-26 05:41:28 | INFO | train rank=0 | step=69440/100000 progress=69.4% loss=0.040100 lr=2.343241e-06 elapsed=25:45:58 avg_step=1.34s recent_step=1.38s eta=11:20:22 +2026-08-26 05:41:43 | INFO | train rank=0 | step=69450/100000 progress=69.5% loss=0.038803 lr=2.341840e-06 elapsed=25:46:13 avg_step=1.34s recent_step=1.48s eta=11:20:09 +2026-08-26 05:41:54 | INFO | train rank=0 | step=69460/100000 progress=69.5% loss=0.054077 lr=2.340440e-06 elapsed=25:46:24 avg_step=1.34s recent_step=1.16s eta=11:19:55 +2026-08-26 05:42:09 | INFO | train rank=0 | step=69470/100000 progress=69.5% loss=0.037140 lr=2.339040e-06 elapsed=25:46:39 avg_step=1.34s recent_step=1.46s eta=11:19:42 +2026-08-26 05:42:24 | INFO | train rank=0 | step=69480/100000 progress=69.5% loss=0.073944 lr=2.337640e-06 elapsed=25:46:53 avg_step=1.34s recent_step=1.47s eta=11:19:29 +2026-08-26 05:42:40 | INFO | train rank=0 | step=69490/100000 progress=69.5% loss=0.036880 lr=2.336240e-06 elapsed=25:47:10 avg_step=1.34s recent_step=1.61s eta=11:19:17 +2026-08-26 05:42:51 | INFO | train rank=0 | step=69500/100000 progress=69.5% loss=0.049225 lr=2.334841e-06 elapsed=25:47:21 avg_step=1.34s recent_step=1.14s eta=11:19:03 +2026-08-26 05:43:03 | INFO | train rank=0 | step=69510/100000 progress=69.5% loss=0.048157 lr=2.333443e-06 elapsed=25:47:33 avg_step=1.34s recent_step=1.19s eta=11:18:49 +2026-08-26 05:43:18 | INFO | train rank=0 | step=69520/100000 progress=69.5% loss=0.044342 lr=2.332044e-06 elapsed=25:47:48 avg_step=1.34s recent_step=1.51s eta=11:18:36 +2026-08-26 05:43:31 | INFO | train rank=0 | step=69530/100000 progress=69.5% loss=0.032372 lr=2.330646e-06 elapsed=25:48:01 avg_step=1.34s recent_step=1.33s eta=11:18:23 +2026-08-26 05:43:47 | INFO | train rank=0 | step=69540/100000 progress=69.5% loss=0.030079 lr=2.329248e-06 elapsed=25:48:16 avg_step=1.34s recent_step=1.52s eta=11:18:10 +2026-08-26 05:44:04 | INFO | train rank=0 | step=69550/100000 progress=69.5% loss=0.048721 lr=2.327850e-06 elapsed=25:48:34 avg_step=1.34s recent_step=1.71s eta=11:17:59 +2026-08-26 05:44:15 | INFO | train rank=0 | step=69560/100000 progress=69.6% loss=0.037277 lr=2.326453e-06 elapsed=25:48:45 avg_step=1.34s recent_step=1.14s eta=11:17:44 +2026-08-26 05:44:29 | INFO | train rank=0 | step=69570/100000 progress=69.6% loss=0.068756 lr=2.325056e-06 elapsed=25:48:59 avg_step=1.34s recent_step=1.43s eta=11:17:31 +2026-08-26 05:44:41 | INFO | train rank=0 | step=69580/100000 progress=69.6% loss=0.027611 lr=2.323659e-06 elapsed=25:49:11 avg_step=1.34s recent_step=1.18s eta=11:17:17 +2026-08-26 05:44:53 | INFO | train rank=0 | step=69590/100000 progress=69.6% loss=0.024601 lr=2.322262e-06 elapsed=25:49:23 avg_step=1.34s recent_step=1.20s eta=11:17:03 +2026-08-26 05:45:05 | INFO | train rank=0 | step=69600/100000 progress=69.6% loss=0.040161 lr=2.320866e-06 elapsed=25:49:35 avg_step=1.34s recent_step=1.16s eta=11:16:49 +2026-08-26 05:45:20 | INFO | train rank=0 | step=69610/100000 progress=69.6% loss=0.080383 lr=2.319470e-06 elapsed=25:49:50 avg_step=1.34s recent_step=1.52s eta=11:16:37 +2026-08-26 05:45:31 | INFO | train rank=0 | step=69620/100000 progress=69.6% loss=0.042908 lr=2.318074e-06 elapsed=25:50:01 avg_step=1.34s recent_step=1.16s eta=11:16:23 +2026-08-26 05:45:45 | INFO | train rank=0 | step=69630/100000 progress=69.6% loss=0.041718 lr=2.316679e-06 elapsed=25:50:15 avg_step=1.34s recent_step=1.33s eta=11:16:09 +2026-08-26 05:45:57 | INFO | train rank=0 | step=69640/100000 progress=69.6% loss=0.028515 lr=2.315284e-06 elapsed=25:50:27 avg_step=1.34s recent_step=1.19s eta=11:15:55 +2026-08-26 05:46:08 | INFO | train rank=0 | step=69650/100000 progress=69.7% loss=0.049896 lr=2.313889e-06 elapsed=25:50:38 avg_step=1.34s recent_step=1.16s eta=11:15:41 +2026-08-26 05:46:25 | INFO | train rank=0 | step=69660/100000 progress=69.7% loss=0.052185 lr=2.312495e-06 elapsed=25:50:55 avg_step=1.34s recent_step=1.70s eta=11:15:29 +2026-08-26 05:46:44 | INFO | train rank=0 | step=69670/100000 progress=69.7% loss=0.041016 lr=2.311101e-06 elapsed=25:51:14 avg_step=1.34s recent_step=1.89s eta=11:15:18 +2026-08-26 05:46:55 | INFO | train rank=0 | step=69680/100000 progress=69.7% loss=0.027847 lr=2.309707e-06 elapsed=25:51:25 avg_step=1.34s recent_step=1.11s eta=11:15:04 +2026-08-26 05:47:07 | INFO | train rank=0 | step=69690/100000 progress=69.7% loss=0.051132 lr=2.308313e-06 elapsed=25:51:37 avg_step=1.34s recent_step=1.18s eta=11:14:50 +2026-08-26 05:47:19 | INFO | train rank=0 | step=69700/100000 progress=69.7% loss=0.047592 lr=2.306920e-06 elapsed=25:51:49 avg_step=1.34s recent_step=1.23s eta=11:14:36 +2026-08-26 05:47:30 | INFO | train rank=0 | step=69710/100000 progress=69.7% loss=0.042160 lr=2.305527e-06 elapsed=25:52:00 avg_step=1.34s recent_step=1.11s eta=11:14:22 +2026-08-26 05:47:45 | INFO | train rank=0 | step=69720/100000 progress=69.7% loss=0.044815 lr=2.304134e-06 elapsed=25:52:15 avg_step=1.34s recent_step=1.49s eta=11:14:09 +2026-08-26 05:47:57 | INFO | train rank=0 | step=69730/100000 progress=69.7% loss=0.063232 lr=2.302742e-06 elapsed=25:52:27 avg_step=1.34s recent_step=1.18s eta=11:13:55 +2026-08-26 05:48:10 | INFO | train rank=0 | step=69740/100000 progress=69.7% loss=0.038498 lr=2.301350e-06 elapsed=25:52:40 avg_step=1.34s recent_step=1.34s eta=11:13:42 +2026-08-26 05:48:22 | INFO | train rank=0 | step=69750/100000 progress=69.8% loss=0.043625 lr=2.299958e-06 elapsed=25:52:52 avg_step=1.34s recent_step=1.20s eta=11:13:28 +2026-08-26 05:48:38 | INFO | train rank=0 | step=69760/100000 progress=69.8% loss=0.055084 lr=2.298566e-06 elapsed=25:53:08 avg_step=1.34s recent_step=1.54s eta=11:13:15 +2026-08-26 05:48:52 | INFO | train rank=0 | step=69770/100000 progress=69.8% loss=0.053970 lr=2.297175e-06 elapsed=25:53:22 avg_step=1.34s recent_step=1.47s eta=11:13:02 +2026-08-26 05:49:07 | INFO | train rank=0 | step=69780/100000 progress=69.8% loss=0.034065 lr=2.295784e-06 elapsed=25:53:37 avg_step=1.34s recent_step=1.48s eta=11:12:50 +2026-08-26 05:49:22 | INFO | train rank=0 | step=69790/100000 progress=69.8% loss=0.047958 lr=2.294394e-06 elapsed=25:53:52 avg_step=1.34s recent_step=1.48s eta=11:12:37 +2026-08-26 05:49:37 | INFO | train rank=0 | step=69800/100000 progress=69.8% loss=0.039474 lr=2.293003e-06 elapsed=25:54:07 avg_step=1.34s recent_step=1.55s eta=11:12:25 +2026-08-26 05:49:49 | INFO | train rank=0 | step=69810/100000 progress=69.8% loss=0.037354 lr=2.291613e-06 elapsed=25:54:18 avg_step=1.34s recent_step=1.11s eta=11:12:10 +2026-08-26 05:50:05 | INFO | train rank=0 | step=69820/100000 progress=69.8% loss=0.050568 lr=2.290224e-06 elapsed=25:54:34 avg_step=1.34s recent_step=1.59s eta=11:11:58 +2026-08-26 05:50:16 | INFO | train rank=0 | step=69830/100000 progress=69.8% loss=0.026947 lr=2.288834e-06 elapsed=25:54:46 avg_step=1.34s recent_step=1.19s eta=11:11:44 +2026-08-26 05:50:28 | INFO | train rank=0 | step=69840/100000 progress=69.8% loss=0.082947 lr=2.287445e-06 elapsed=25:54:57 avg_step=1.34s recent_step=1.11s eta=11:11:30 +2026-08-26 05:50:41 | INFO | train rank=0 | step=69850/100000 progress=69.8% loss=0.054092 lr=2.286056e-06 elapsed=25:55:11 avg_step=1.34s recent_step=1.35s eta=11:11:16 +2026-08-26 05:50:53 | INFO | train rank=0 | step=69860/100000 progress=69.9% loss=0.060608 lr=2.284668e-06 elapsed=25:55:22 avg_step=1.34s recent_step=1.15s eta=11:11:02 +2026-08-26 05:51:06 | INFO | train rank=0 | step=69870/100000 progress=69.9% loss=0.049927 lr=2.283279e-06 elapsed=25:55:36 avg_step=1.34s recent_step=1.37s eta=11:10:49 +2026-08-26 05:51:21 | INFO | train rank=0 | step=69880/100000 progress=69.9% loss=0.042885 lr=2.281891e-06 elapsed=25:55:51 avg_step=1.34s recent_step=1.50s eta=11:10:36 +2026-08-26 05:51:33 | INFO | train rank=0 | step=69890/100000 progress=69.9% loss=0.042175 lr=2.280504e-06 elapsed=25:56:03 avg_step=1.34s recent_step=1.16s eta=11:10:22 +2026-08-26 05:51:48 | INFO | train rank=0 | step=69900/100000 progress=69.9% loss=0.039688 lr=2.279116e-06 elapsed=25:56:18 avg_step=1.34s recent_step=1.49s eta=11:10:10 +2026-08-26 05:52:04 | INFO | train rank=0 | step=69910/100000 progress=69.9% loss=0.037445 lr=2.277729e-06 elapsed=25:56:33 avg_step=1.34s recent_step=1.57s eta=11:09:57 +2026-08-26 05:52:15 | INFO | train rank=0 | step=69920/100000 progress=69.9% loss=0.043274 lr=2.276343e-06 elapsed=25:56:45 avg_step=1.34s recent_step=1.17s eta=11:09:43 +2026-08-26 05:52:31 | INFO | train rank=0 | step=69930/100000 progress=69.9% loss=0.064178 lr=2.274956e-06 elapsed=25:57:01 avg_step=1.34s recent_step=1.62s eta=11:09:31 +2026-08-26 05:52:43 | INFO | train rank=0 | step=69940/100000 progress=69.9% loss=0.067383 lr=2.273570e-06 elapsed=25:57:13 avg_step=1.34s recent_step=1.15s eta=11:09:17 +2026-08-26 05:52:57 | INFO | train rank=0 | step=69950/100000 progress=70.0% loss=0.039673 lr=2.272184e-06 elapsed=25:57:26 avg_step=1.34s recent_step=1.36s eta=11:09:04 +2026-08-26 05:53:09 | INFO | train rank=0 | step=69960/100000 progress=70.0% loss=0.050415 lr=2.270799e-06 elapsed=25:57:38 avg_step=1.34s recent_step=1.20s eta=11:08:50 +2026-08-26 05:53:23 | INFO | train rank=0 | step=69970/100000 progress=70.0% loss=0.042305 lr=2.269413e-06 elapsed=25:57:53 avg_step=1.34s recent_step=1.43s eta=11:08:37 +2026-08-26 05:53:36 | INFO | train rank=0 | step=69980/100000 progress=70.0% loss=0.033020 lr=2.268028e-06 elapsed=25:58:06 avg_step=1.34s recent_step=1.28s eta=11:08:23 +2026-08-26 05:53:48 | INFO | train rank=0 | step=69990/100000 progress=70.0% loss=0.028496 lr=2.266644e-06 elapsed=25:58:18 avg_step=1.34s recent_step=1.20s eta=11:08:09 +2026-08-26 05:54:00 | INFO | train rank=0 | step=70000/100000 progress=70.0% loss=0.054619 lr=2.265259e-06 elapsed=25:58:29 avg_step=1.34s recent_step=1.19s eta=11:07:55 +2026-08-26 05:54:13 | INFO | train rank=0 | step=70010/100000 progress=70.0% loss=0.059845 lr=2.263875e-06 elapsed=25:58:42 avg_step=1.34s recent_step=1.30s eta=11:07:42 +2026-08-26 05:54:27 | INFO | train rank=0 | step=70020/100000 progress=70.0% loss=0.054810 lr=2.262491e-06 elapsed=25:58:57 avg_step=1.34s recent_step=1.47s eta=11:07:29 +2026-08-26 05:54:40 | INFO | train rank=0 | step=70030/100000 progress=70.0% loss=0.051773 lr=2.261108e-06 elapsed=25:59:10 avg_step=1.34s recent_step=1.29s eta=11:07:15 +2026-08-26 05:54:59 | INFO | train rank=0 | step=70040/100000 progress=70.0% loss=0.038940 lr=2.259725e-06 elapsed=25:59:29 avg_step=1.34s recent_step=1.86s eta=11:07:04 +2026-08-26 05:55:10 | INFO | train rank=0 | step=70050/100000 progress=70.0% loss=0.046608 lr=2.258342e-06 elapsed=25:59:40 avg_step=1.34s recent_step=1.14s eta=11:06:50 +2026-08-26 05:55:23 | INFO | train rank=0 | step=70060/100000 progress=70.1% loss=0.061035 lr=2.256959e-06 elapsed=25:59:53 avg_step=1.34s recent_step=1.31s eta=11:06:37 +2026-08-26 05:55:35 | INFO | train rank=0 | step=70070/100000 progress=70.1% loss=0.039490 lr=2.255577e-06 elapsed=26:00:05 avg_step=1.34s recent_step=1.22s eta=11:06:23 +2026-08-26 05:55:49 | INFO | train rank=0 | step=70080/100000 progress=70.1% loss=0.061859 lr=2.254195e-06 elapsed=26:00:19 avg_step=1.34s recent_step=1.39s eta=11:06:10 +2026-08-26 05:56:05 | INFO | train rank=0 | step=70090/100000 progress=70.1% loss=0.056519 lr=2.252813e-06 elapsed=26:00:35 avg_step=1.34s recent_step=1.55s eta=11:05:57 +2026-08-26 05:56:16 | INFO | train rank=0 | step=70100/100000 progress=70.1% loss=0.051224 lr=2.251432e-06 elapsed=26:00:46 avg_step=1.34s recent_step=1.12s eta=11:05:43 +2026-08-26 05:56:28 | INFO | train rank=0 | step=70110/100000 progress=70.1% loss=0.075287 lr=2.250051e-06 elapsed=26:00:58 avg_step=1.34s recent_step=1.20s eta=11:05:29 +2026-08-26 05:56:47 | INFO | train rank=0 | step=70120/100000 progress=70.1% loss=0.050888 lr=2.248670e-06 elapsed=26:01:17 avg_step=1.34s recent_step=1.87s eta=11:05:18 +2026-08-26 05:56:59 | INFO | train rank=0 | step=70130/100000 progress=70.1% loss=0.074005 lr=2.247290e-06 elapsed=26:01:29 avg_step=1.34s recent_step=1.19s eta=11:05:04 +2026-08-26 05:57:10 | INFO | train rank=0 | step=70140/100000 progress=70.1% loss=0.111084 lr=2.245909e-06 elapsed=26:01:40 avg_step=1.34s recent_step=1.15s eta=11:04:50 +2026-08-26 05:57:22 | INFO | train rank=0 | step=70150/100000 progress=70.2% loss=0.051697 lr=2.244529e-06 elapsed=26:01:52 avg_step=1.34s recent_step=1.15s eta=11:04:36 +2026-08-26 05:57:37 | INFO | train rank=0 | step=70160/100000 progress=70.2% loss=0.043564 lr=2.243150e-06 elapsed=26:02:06 avg_step=1.34s recent_step=1.49s eta=11:04:23 +2026-08-26 05:57:49 | INFO | train rank=0 | step=70170/100000 progress=70.2% loss=0.046432 lr=2.241771e-06 elapsed=26:02:19 avg_step=1.34s recent_step=1.21s eta=11:04:09 +2026-08-26 05:58:00 | INFO | train rank=0 | step=70180/100000 progress=70.2% loss=0.054794 lr=2.240392e-06 elapsed=26:02:30 avg_step=1.34s recent_step=1.18s eta=11:03:55 +2026-08-26 05:58:13 | INFO | train rank=0 | step=70190/100000 progress=70.2% loss=0.038071 lr=2.239013e-06 elapsed=26:02:42 avg_step=1.34s recent_step=1.21s eta=11:03:41 +2026-08-26 05:58:28 | INFO | train rank=0 | step=70200/100000 progress=70.2% loss=0.034149 lr=2.237635e-06 elapsed=26:02:58 avg_step=1.34s recent_step=1.51s eta=11:03:28 +2026-08-26 05:58:42 | INFO | train rank=0 | step=70210/100000 progress=70.2% loss=0.051636 lr=2.236257e-06 elapsed=26:03:12 avg_step=1.34s recent_step=1.40s eta=11:03:15 +2026-08-26 05:58:54 | INFO | train rank=0 | step=70220/100000 progress=70.2% loss=0.044693 lr=2.234879e-06 elapsed=26:03:24 avg_step=1.34s recent_step=1.26s eta=11:03:02 +2026-08-26 05:59:09 | INFO | train rank=0 | step=70230/100000 progress=70.2% loss=0.068329 lr=2.233501e-06 elapsed=26:03:39 avg_step=1.34s recent_step=1.51s eta=11:02:49 +2026-08-26 05:59:21 | INFO | train rank=0 | step=70240/100000 progress=70.2% loss=0.042542 lr=2.232124e-06 elapsed=26:03:51 avg_step=1.34s recent_step=1.17s eta=11:02:35 +2026-08-26 05:59:36 | INFO | train rank=0 | step=70250/100000 progress=70.2% loss=0.060608 lr=2.230747e-06 elapsed=26:04:06 avg_step=1.34s recent_step=1.51s eta=11:02:22 +2026-08-26 05:59:48 | INFO | train rank=0 | step=70260/100000 progress=70.3% loss=0.038818 lr=2.229371e-06 elapsed=26:04:18 avg_step=1.34s recent_step=1.14s eta=11:02:08 +2026-08-26 06:00:03 | INFO | train rank=0 | step=70270/100000 progress=70.3% loss=0.067932 lr=2.227995e-06 elapsed=26:04:33 avg_step=1.34s recent_step=1.57s eta=11:01:56 +2026-08-26 06:00:15 | INFO | train rank=0 | step=70280/100000 progress=70.3% loss=0.051468 lr=2.226619e-06 elapsed=26:04:45 avg_step=1.34s recent_step=1.16s eta=11:01:42 +2026-08-26 06:00:27 | INFO | train rank=0 | step=70290/100000 progress=70.3% loss=0.050934 lr=2.225243e-06 elapsed=26:04:57 avg_step=1.34s recent_step=1.25s eta=11:01:28 +2026-08-26 06:00:45 | INFO | train rank=0 | step=70300/100000 progress=70.3% loss=0.035156 lr=2.223868e-06 elapsed=26:05:14 avg_step=1.34s recent_step=1.72s eta=11:01:16 +2026-08-26 06:01:00 | INFO | train rank=0 | step=70310/100000 progress=70.3% loss=0.039078 lr=2.222493e-06 elapsed=26:05:30 avg_step=1.34s recent_step=1.59s eta=11:01:04 +2026-08-26 06:01:13 | INFO | train rank=0 | step=70320/100000 progress=70.3% loss=0.070984 lr=2.221118e-06 elapsed=26:05:42 avg_step=1.34s recent_step=1.21s eta=11:00:50 +2026-08-26 06:01:29 | INFO | train rank=0 | step=70330/100000 progress=70.3% loss=0.072510 lr=2.219743e-06 elapsed=26:05:59 avg_step=1.34s recent_step=1.66s eta=11:00:38 +2026-08-26 06:01:41 | INFO | train rank=0 | step=70340/100000 progress=70.3% loss=0.049622 lr=2.218369e-06 elapsed=26:06:11 avg_step=1.34s recent_step=1.20s eta=11:00:24 +2026-08-26 06:01:53 | INFO | train rank=0 | step=70350/100000 progress=70.3% loss=0.043518 lr=2.216995e-06 elapsed=26:06:23 avg_step=1.34s recent_step=1.16s eta=11:00:10 +2026-08-26 06:02:04 | INFO | train rank=0 | step=70360/100000 progress=70.4% loss=0.046509 lr=2.215622e-06 elapsed=26:06:34 avg_step=1.34s recent_step=1.15s eta=10:59:56 +2026-08-26 06:02:16 | INFO | train rank=0 | step=70370/100000 progress=70.4% loss=0.050705 lr=2.214249e-06 elapsed=26:06:46 avg_step=1.34s recent_step=1.20s eta=10:59:42 +2026-08-26 06:02:30 | INFO | train rank=0 | step=70380/100000 progress=70.4% loss=0.053711 lr=2.212876e-06 elapsed=26:07:00 avg_step=1.34s recent_step=1.38s eta=10:59:29 +2026-08-26 06:02:42 | INFO | train rank=0 | step=70390/100000 progress=70.4% loss=0.066673 lr=2.211503e-06 elapsed=26:07:12 avg_step=1.34s recent_step=1.19s eta=10:59:15 +2026-08-26 06:02:54 | INFO | train rank=0 | step=70400/100000 progress=70.4% loss=0.049103 lr=2.210131e-06 elapsed=26:07:24 avg_step=1.34s recent_step=1.18s eta=10:59:01 +2026-08-26 06:03:07 | INFO | train rank=0 | step=70410/100000 progress=70.4% loss=0.034615 lr=2.208759e-06 elapsed=26:07:37 avg_step=1.34s recent_step=1.34s eta=10:58:48 +2026-08-26 06:03:22 | INFO | train rank=0 | step=70420/100000 progress=70.4% loss=0.066223 lr=2.207387e-06 elapsed=26:07:52 avg_step=1.34s recent_step=1.51s eta=10:58:35 +2026-08-26 06:03:35 | INFO | train rank=0 | step=70430/100000 progress=70.4% loss=0.041046 lr=2.206016e-06 elapsed=26:08:04 avg_step=1.34s recent_step=1.21s eta=10:58:21 +2026-08-26 06:03:53 | INFO | train rank=0 | step=70440/100000 progress=70.4% loss=0.057861 lr=2.204645e-06 elapsed=26:08:23 avg_step=1.34s recent_step=1.82s eta=10:58:10 +2026-08-26 06:04:09 | INFO | train rank=0 | step=70450/100000 progress=70.5% loss=0.058502 lr=2.203274e-06 elapsed=26:08:38 avg_step=1.34s recent_step=1.58s eta=10:57:57 +2026-08-26 06:04:20 | INFO | train rank=0 | step=70460/100000 progress=70.5% loss=0.039337 lr=2.201904e-06 elapsed=26:08:50 avg_step=1.34s recent_step=1.19s eta=10:57:43 +2026-08-26 06:04:32 | INFO | train rank=0 | step=70470/100000 progress=70.5% loss=0.056717 lr=2.200533e-06 elapsed=26:09:02 avg_step=1.34s recent_step=1.17s eta=10:57:29 +2026-08-26 06:04:44 | INFO | train rank=0 | step=70480/100000 progress=70.5% loss=0.037910 lr=2.199164e-06 elapsed=26:09:14 avg_step=1.34s recent_step=1.15s eta=10:57:15 +2026-08-26 06:04:58 | INFO | train rank=0 | step=70490/100000 progress=70.5% loss=0.043503 lr=2.197794e-06 elapsed=26:09:28 avg_step=1.34s recent_step=1.47s eta=10:57:02 +2026-08-26 06:05:11 | INFO | train rank=0 | step=70500/100000 progress=70.5% loss=0.035889 lr=2.196425e-06 elapsed=26:09:41 avg_step=1.34s recent_step=1.24s eta=10:56:49 +2026-08-26 06:05:26 | INFO | train rank=0 | step=70510/100000 progress=70.5% loss=0.040314 lr=2.195056e-06 elapsed=26:09:56 avg_step=1.34s recent_step=1.56s eta=10:56:36 +2026-08-26 06:05:42 | INFO | train rank=0 | step=70520/100000 progress=70.5% loss=0.041779 lr=2.193687e-06 elapsed=26:10:12 avg_step=1.34s recent_step=1.57s eta=10:56:24 +2026-08-26 06:05:54 | INFO | train rank=0 | step=70530/100000 progress=70.5% loss=0.045059 lr=2.192319e-06 elapsed=26:10:24 avg_step=1.34s recent_step=1.20s eta=10:56:10 +2026-08-26 06:06:07 | INFO | train rank=0 | step=70540/100000 progress=70.5% loss=0.052948 lr=2.190951e-06 elapsed=26:10:37 avg_step=1.34s recent_step=1.33s eta=10:55:56 +2026-08-26 06:06:19 | INFO | train rank=0 | step=70550/100000 progress=70.5% loss=0.046455 lr=2.189583e-06 elapsed=26:10:49 avg_step=1.34s recent_step=1.17s eta=10:55:42 +2026-08-26 06:06:35 | INFO | train rank=0 | step=70560/100000 progress=70.6% loss=0.055023 lr=2.188216e-06 elapsed=26:11:05 avg_step=1.34s recent_step=1.62s eta=10:55:30 +2026-08-26 06:06:46 | INFO | train rank=0 | step=70570/100000 progress=70.6% loss=0.036232 lr=2.186849e-06 elapsed=26:11:16 avg_step=1.34s recent_step=1.12s eta=10:55:16 +2026-08-26 06:07:04 | INFO | train rank=0 | step=70580/100000 progress=70.6% loss=0.083038 lr=2.185482e-06 elapsed=26:11:34 avg_step=1.34s recent_step=1.82s eta=10:55:05 +2026-08-26 06:07:19 | INFO | train rank=0 | step=70590/100000 progress=70.6% loss=0.050323 lr=2.184115e-06 elapsed=26:11:49 avg_step=1.34s recent_step=1.43s eta=10:54:52 +2026-08-26 06:07:32 | INFO | train rank=0 | step=70600/100000 progress=70.6% loss=0.034256 lr=2.182749e-06 elapsed=26:12:02 avg_step=1.34s recent_step=1.32s eta=10:54:38 +2026-08-26 06:07:47 | INFO | train rank=0 | step=70610/100000 progress=70.6% loss=0.052307 lr=2.181383e-06 elapsed=26:12:17 avg_step=1.34s recent_step=1.48s eta=10:54:25 +2026-08-26 06:08:00 | INFO | train rank=0 | step=70620/100000 progress=70.6% loss=0.055176 lr=2.180018e-06 elapsed=26:12:30 avg_step=1.34s recent_step=1.32s eta=10:54:12 +2026-08-26 06:08:12 | INFO | train rank=0 | step=70630/100000 progress=70.6% loss=0.032562 lr=2.178653e-06 elapsed=26:12:41 avg_step=1.34s recent_step=1.16s eta=10:53:58 +2026-08-26 06:08:23 | INFO | train rank=0 | step=70640/100000 progress=70.6% loss=0.046249 lr=2.177288e-06 elapsed=26:12:53 avg_step=1.34s recent_step=1.19s eta=10:53:44 +2026-08-26 06:08:38 | INFO | train rank=0 | step=70650/100000 progress=70.7% loss=0.062805 lr=2.175923e-06 elapsed=26:13:08 avg_step=1.34s recent_step=1.45s eta=10:53:31 +2026-08-26 06:08:53 | INFO | train rank=0 | step=70660/100000 progress=70.7% loss=0.065033 lr=2.174559e-06 elapsed=26:13:23 avg_step=1.34s recent_step=1.48s eta=10:53:18 +2026-08-26 06:09:04 | INFO | train rank=0 | step=70670/100000 progress=70.7% loss=0.034943 lr=2.173195e-06 elapsed=26:13:34 avg_step=1.34s recent_step=1.17s eta=10:53:04 +2026-08-26 06:09:16 | INFO | train rank=0 | step=70680/100000 progress=70.7% loss=0.065308 lr=2.171831e-06 elapsed=26:13:45 avg_step=1.34s recent_step=1.10s eta=10:52:50 +2026-08-26 06:09:29 | INFO | train rank=0 | step=70690/100000 progress=70.7% loss=0.035133 lr=2.170468e-06 elapsed=26:13:59 avg_step=1.34s recent_step=1.34s eta=10:52:37 +2026-08-26 06:09:44 | INFO | train rank=0 | step=70700/100000 progress=70.7% loss=0.079010 lr=2.169104e-06 elapsed=26:14:14 avg_step=1.34s recent_step=1.47s eta=10:52:24 +2026-08-26 06:09:58 | INFO | train rank=0 | step=70710/100000 progress=70.7% loss=0.062988 lr=2.167742e-06 elapsed=26:14:28 avg_step=1.34s recent_step=1.43s eta=10:52:11 +2026-08-26 06:10:10 | INFO | train rank=0 | step=70720/100000 progress=70.7% loss=0.035675 lr=2.166379e-06 elapsed=26:14:40 avg_step=1.34s recent_step=1.19s eta=10:51:57 +2026-08-26 06:10:21 | INFO | train rank=0 | step=70730/100000 progress=70.7% loss=0.046288 lr=2.165017e-06 elapsed=26:14:51 avg_step=1.34s recent_step=1.13s eta=10:51:43 +2026-08-26 06:10:38 | INFO | train rank=0 | step=70740/100000 progress=70.7% loss=0.037582 lr=2.163655e-06 elapsed=26:15:08 avg_step=1.34s recent_step=1.69s eta=10:51:31 +2026-08-26 06:10:50 | INFO | train rank=0 | step=70750/100000 progress=70.8% loss=0.039925 lr=2.162294e-06 elapsed=26:15:20 avg_step=1.34s recent_step=1.23s eta=10:51:17 +2026-08-26 06:11:02 | INFO | train rank=0 | step=70760/100000 progress=70.8% loss=0.057556 lr=2.160933e-06 elapsed=26:15:32 avg_step=1.34s recent_step=1.15s eta=10:51:03 +2026-08-26 06:11:17 | INFO | train rank=0 | step=70770/100000 progress=70.8% loss=0.045929 lr=2.159572e-06 elapsed=26:15:47 avg_step=1.34s recent_step=1.55s eta=10:50:50 +2026-08-26 06:11:29 | INFO | train rank=0 | step=70780/100000 progress=70.8% loss=0.027962 lr=2.158211e-06 elapsed=26:15:59 avg_step=1.34s recent_step=1.16s eta=10:50:36 +2026-08-26 06:11:43 | INFO | train rank=0 | step=70790/100000 progress=70.8% loss=0.065704 lr=2.156851e-06 elapsed=26:16:13 avg_step=1.34s recent_step=1.40s eta=10:50:23 +2026-08-26 06:11:54 | INFO | train rank=0 | step=70800/100000 progress=70.8% loss=0.042160 lr=2.155491e-06 elapsed=26:16:24 avg_step=1.34s recent_step=1.15s eta=10:50:09 +2026-08-26 06:12:12 | INFO | train rank=0 | step=70810/100000 progress=70.8% loss=0.049911 lr=2.154131e-06 elapsed=26:16:42 avg_step=1.34s recent_step=1.77s eta=10:49:57 +2026-08-26 06:12:27 | INFO | train rank=0 | step=70820/100000 progress=70.8% loss=0.033112 lr=2.152772e-06 elapsed=26:16:57 avg_step=1.34s recent_step=1.47s eta=10:49:45 +2026-08-26 06:12:43 | INFO | train rank=0 | step=70830/100000 progress=70.8% loss=0.095612 lr=2.151413e-06 elapsed=26:17:12 avg_step=1.34s recent_step=1.58s eta=10:49:32 +2026-08-26 06:12:56 | INFO | train rank=0 | step=70840/100000 progress=70.8% loss=0.030777 lr=2.150054e-06 elapsed=26:17:26 avg_step=1.34s recent_step=1.32s eta=10:49:19 +2026-08-26 06:13:10 | INFO | train rank=0 | step=70850/100000 progress=70.9% loss=0.036713 lr=2.148696e-06 elapsed=26:17:40 avg_step=1.34s recent_step=1.47s eta=10:49:06 +2026-08-26 06:13:24 | INFO | train rank=0 | step=70860/100000 progress=70.9% loss=0.049454 lr=2.147337e-06 elapsed=26:17:54 avg_step=1.34s recent_step=1.34s eta=10:48:53 +2026-08-26 06:13:36 | INFO | train rank=0 | step=70870/100000 progress=70.9% loss=0.059998 lr=2.145980e-06 elapsed=26:18:05 avg_step=1.34s recent_step=1.17s eta=10:48:39 +2026-08-26 06:13:50 | INFO | train rank=0 | step=70880/100000 progress=70.9% loss=0.056671 lr=2.144622e-06 elapsed=26:18:20 avg_step=1.34s recent_step=1.45s eta=10:48:26 +2026-08-26 06:14:04 | INFO | train rank=0 | step=70890/100000 progress=70.9% loss=0.043854 lr=2.143265e-06 elapsed=26:18:34 avg_step=1.34s recent_step=1.43s eta=10:48:13 +2026-08-26 06:14:16 | INFO | train rank=0 | step=70900/100000 progress=70.9% loss=0.050301 lr=2.141908e-06 elapsed=26:18:46 avg_step=1.34s recent_step=1.19s eta=10:47:59 +2026-08-26 06:14:33 | INFO | train rank=0 | step=70910/100000 progress=70.9% loss=0.049347 lr=2.140552e-06 elapsed=26:19:03 avg_step=1.34s recent_step=1.69s eta=10:47:47 +2026-08-26 06:14:45 | INFO | train rank=0 | step=70920/100000 progress=70.9% loss=0.044250 lr=2.139195e-06 elapsed=26:19:14 avg_step=1.34s recent_step=1.16s eta=10:47:33 +2026-08-26 06:14:57 | INFO | train rank=0 | step=70930/100000 progress=70.9% loss=0.059052 lr=2.137839e-06 elapsed=26:19:26 avg_step=1.34s recent_step=1.20s eta=10:47:19 +2026-08-26 06:15:10 | INFO | train rank=0 | step=70940/100000 progress=70.9% loss=0.062500 lr=2.136484e-06 elapsed=26:19:40 avg_step=1.34s recent_step=1.35s eta=10:47:06 +2026-08-26 06:15:21 | INFO | train rank=0 | step=70950/100000 progress=71.0% loss=0.061401 lr=2.135128e-06 elapsed=26:19:51 avg_step=1.34s recent_step=1.11s eta=10:46:51 +2026-08-26 06:15:41 | INFO | train rank=0 | step=70960/100000 progress=71.0% loss=0.029530 lr=2.133773e-06 elapsed=26:20:11 avg_step=1.34s recent_step=2.02s eta=10:46:41 +2026-08-26 06:15:52 | INFO | train rank=0 | step=70970/100000 progress=71.0% loss=0.038681 lr=2.132419e-06 elapsed=26:20:22 avg_step=1.34s recent_step=1.09s eta=10:46:26 +2026-08-26 06:16:04 | INFO | train rank=0 | step=70980/100000 progress=71.0% loss=0.043243 lr=2.131064e-06 elapsed=26:20:34 avg_step=1.34s recent_step=1.15s eta=10:46:12 +2026-08-26 06:16:16 | INFO | train rank=0 | step=70990/100000 progress=71.0% loss=0.046265 lr=2.129710e-06 elapsed=26:20:46 avg_step=1.34s recent_step=1.19s eta=10:45:58 +2026-08-26 06:16:27 | INFO | train rank=0 | step=71000/100000 progress=71.0% loss=0.046227 lr=2.128357e-06 elapsed=26:20:57 avg_step=1.34s recent_step=1.14s eta=10:45:44 +2026-08-26 06:16:39 | INFO | train rank=0 | step=71010/100000 progress=71.0% loss=0.039886 lr=2.127003e-06 elapsed=26:21:09 avg_step=1.34s recent_step=1.21s eta=10:45:30 +2026-08-26 06:16:51 | INFO | train rank=0 | step=71020/100000 progress=71.0% loss=0.036156 lr=2.125650e-06 elapsed=26:21:21 avg_step=1.34s recent_step=1.16s eta=10:45:16 +2026-08-26 06:17:02 | INFO | train rank=0 | step=71030/100000 progress=71.0% loss=0.126587 lr=2.124297e-06 elapsed=26:21:32 avg_step=1.34s recent_step=1.17s eta=10:45:02 +2026-08-26 06:17:15 | INFO | train rank=0 | step=71040/100000 progress=71.0% loss=0.073334 lr=2.122945e-06 elapsed=26:21:45 avg_step=1.34s recent_step=1.26s eta=10:44:48 +2026-08-26 06:17:30 | INFO | train rank=0 | step=71050/100000 progress=71.0% loss=0.045349 lr=2.121593e-06 elapsed=26:22:00 avg_step=1.34s recent_step=1.50s eta=10:44:36 +2026-08-26 06:17:44 | INFO | train rank=0 | step=71060/100000 progress=71.1% loss=0.042984 lr=2.120241e-06 elapsed=26:22:14 avg_step=1.34s recent_step=1.41s eta=10:44:23 +2026-08-26 06:17:56 | INFO | train rank=0 | step=71070/100000 progress=71.1% loss=0.034348 lr=2.118889e-06 elapsed=26:22:26 avg_step=1.34s recent_step=1.14s eta=10:44:09 +2026-08-26 06:18:10 | INFO | train rank=0 | step=71080/100000 progress=71.1% loss=0.043869 lr=2.117538e-06 elapsed=26:22:40 avg_step=1.34s recent_step=1.43s eta=10:43:56 +2026-08-26 06:18:25 | INFO | train rank=0 | step=71090/100000 progress=71.1% loss=0.047737 lr=2.116187e-06 elapsed=26:22:55 avg_step=1.34s recent_step=1.51s eta=10:43:43 +2026-08-26 06:18:40 | INFO | train rank=0 | step=71100/100000 progress=71.1% loss=0.045288 lr=2.114837e-06 elapsed=26:23:10 avg_step=1.34s recent_step=1.50s eta=10:43:30 +2026-08-26 06:18:53 | INFO | train rank=0 | step=71110/100000 progress=71.1% loss=0.040695 lr=2.113486e-06 elapsed=26:23:23 avg_step=1.34s recent_step=1.28s eta=10:43:17 +2026-08-26 06:19:05 | INFO | train rank=0 | step=71120/100000 progress=71.1% loss=0.049347 lr=2.112136e-06 elapsed=26:23:35 avg_step=1.34s recent_step=1.21s eta=10:43:03 +2026-08-26 06:19:17 | INFO | train rank=0 | step=71130/100000 progress=71.1% loss=0.051376 lr=2.110787e-06 elapsed=26:23:47 avg_step=1.34s recent_step=1.18s eta=10:42:49 +2026-08-26 06:19:36 | INFO | train rank=0 | step=71140/100000 progress=71.1% loss=0.038605 lr=2.109438e-06 elapsed=26:24:06 avg_step=1.34s recent_step=1.92s eta=10:42:38 +2026-08-26 06:19:51 | INFO | train rank=0 | step=71150/100000 progress=71.2% loss=0.065750 lr=2.108089e-06 elapsed=26:24:21 avg_step=1.34s recent_step=1.48s eta=10:42:25 +2026-08-26 06:20:02 | INFO | train rank=0 | step=71160/100000 progress=71.2% loss=0.057434 lr=2.106740e-06 elapsed=26:24:32 avg_step=1.34s recent_step=1.16s eta=10:42:11 +2026-08-26 06:20:15 | INFO | train rank=0 | step=71170/100000 progress=71.2% loss=0.085175 lr=2.105391e-06 elapsed=26:24:45 avg_step=1.34s recent_step=1.25s eta=10:41:57 +2026-08-26 06:20:29 | INFO | train rank=0 | step=71180/100000 progress=71.2% loss=0.065918 lr=2.104043e-06 elapsed=26:24:59 avg_step=1.34s recent_step=1.45s eta=10:41:44 +2026-08-26 06:20:42 | INFO | train rank=0 | step=71190/100000 progress=71.2% loss=0.052994 lr=2.102696e-06 elapsed=26:25:12 avg_step=1.34s recent_step=1.26s eta=10:41:31 +2026-08-26 06:20:55 | INFO | train rank=0 | step=71200/100000 progress=71.2% loss=0.034904 lr=2.101348e-06 elapsed=26:25:25 avg_step=1.34s recent_step=1.33s eta=10:41:17 +2026-08-26 06:21:09 | INFO | train rank=0 | step=71210/100000 progress=71.2% loss=0.045013 lr=2.100001e-06 elapsed=26:25:39 avg_step=1.34s recent_step=1.35s eta=10:41:04 +2026-08-26 06:21:21 | INFO | train rank=0 | step=71220/100000 progress=71.2% loss=0.048294 lr=2.098654e-06 elapsed=26:25:51 avg_step=1.34s recent_step=1.26s eta=10:40:50 +2026-08-26 06:21:34 | INFO | train rank=0 | step=71230/100000 progress=71.2% loss=0.052277 lr=2.097308e-06 elapsed=26:26:04 avg_step=1.34s recent_step=1.28s eta=10:40:37 +2026-08-26 06:21:47 | INFO | train rank=0 | step=71240/100000 progress=71.2% loss=0.062042 lr=2.095962e-06 elapsed=26:26:17 avg_step=1.34s recent_step=1.31s eta=10:40:23 +2026-08-26 06:22:00 | INFO | train rank=0 | step=71250/100000 progress=71.2% loss=0.047806 lr=2.094616e-06 elapsed=26:26:30 avg_step=1.34s recent_step=1.31s eta=10:40:10 +2026-08-26 06:22:14 | INFO | train rank=0 | step=71260/100000 progress=71.3% loss=0.043953 lr=2.093270e-06 elapsed=26:26:44 avg_step=1.34s recent_step=1.36s eta=10:39:57 +2026-08-26 06:22:26 | INFO | train rank=0 | step=71270/100000 progress=71.3% loss=0.056015 lr=2.091925e-06 elapsed=26:26:56 avg_step=1.34s recent_step=1.22s eta=10:39:43 +2026-08-26 06:22:39 | INFO | train rank=0 | step=71280/100000 progress=71.3% loss=0.054947 lr=2.090580e-06 elapsed=26:27:09 avg_step=1.34s recent_step=1.26s eta=10:39:29 +2026-08-26 06:22:52 | INFO | train rank=0 | step=71290/100000 progress=71.3% loss=0.057648 lr=2.089236e-06 elapsed=26:27:22 avg_step=1.34s recent_step=1.34s eta=10:39:16 +2026-08-26 06:23:06 | INFO | train rank=0 | step=71300/100000 progress=71.3% loss=0.030693 lr=2.087892e-06 elapsed=26:27:36 avg_step=1.34s recent_step=1.36s eta=10:39:02 +2026-08-26 06:23:18 | INFO | train rank=0 | step=71310/100000 progress=71.3% loss=0.056931 lr=2.086548e-06 elapsed=26:27:48 avg_step=1.34s recent_step=1.24s eta=10:38:49 +2026-08-26 06:23:33 | INFO | train rank=0 | step=71320/100000 progress=71.3% loss=0.047028 lr=2.085204e-06 elapsed=26:28:03 avg_step=1.34s recent_step=1.48s eta=10:38:36 +2026-08-26 06:23:50 | INFO | train rank=0 | step=71330/100000 progress=71.3% loss=0.033722 lr=2.083861e-06 elapsed=26:28:20 avg_step=1.34s recent_step=1.72s eta=10:38:24 +2026-08-26 06:24:08 | INFO | train rank=0 | step=71340/100000 progress=71.3% loss=0.043579 lr=2.082518e-06 elapsed=26:28:38 avg_step=1.34s recent_step=1.75s eta=10:38:12 +2026-08-26 06:24:20 | INFO | train rank=0 | step=71350/100000 progress=71.4% loss=0.033775 lr=2.081175e-06 elapsed=26:28:50 avg_step=1.34s recent_step=1.21s eta=10:37:59 +2026-08-26 06:24:33 | INFO | train rank=0 | step=71360/100000 progress=71.4% loss=0.041740 lr=2.079833e-06 elapsed=26:29:02 avg_step=1.34s recent_step=1.27s eta=10:37:45 +2026-08-26 06:24:46 | INFO | train rank=0 | step=71370/100000 progress=71.4% loss=0.044312 lr=2.078491e-06 elapsed=26:29:16 avg_step=1.34s recent_step=1.36s eta=10:37:32 +2026-08-26 06:24:59 | INFO | train rank=0 | step=71380/100000 progress=71.4% loss=0.050415 lr=2.077149e-06 elapsed=26:29:29 avg_step=1.34s recent_step=1.25s eta=10:37:18 +2026-08-26 06:25:11 | INFO | train rank=0 | step=71390/100000 progress=71.4% loss=0.029137 lr=2.075808e-06 elapsed=26:29:41 avg_step=1.34s recent_step=1.23s eta=10:37:04 +2026-08-26 06:25:24 | INFO | train rank=0 | step=71400/100000 progress=71.4% loss=0.119385 lr=2.074467e-06 elapsed=26:29:54 avg_step=1.34s recent_step=1.27s eta=10:36:51 +2026-08-26 06:25:37 | INFO | train rank=0 | step=71410/100000 progress=71.4% loss=0.057861 lr=2.073126e-06 elapsed=26:30:07 avg_step=1.34s recent_step=1.36s eta=10:36:37 +2026-08-26 06:25:49 | INFO | train rank=0 | step=71420/100000 progress=71.4% loss=0.088623 lr=2.071785e-06 elapsed=26:30:19 avg_step=1.34s recent_step=1.17s eta=10:36:23 +2026-08-26 06:26:02 | INFO | train rank=0 | step=71430/100000 progress=71.4% loss=0.037888 lr=2.070445e-06 elapsed=26:30:32 avg_step=1.34s recent_step=1.26s eta=10:36:10 +2026-08-26 06:26:14 | INFO | train rank=0 | step=71440/100000 progress=71.4% loss=0.046265 lr=2.069106e-06 elapsed=26:30:44 avg_step=1.34s recent_step=1.21s eta=10:35:56 +2026-08-26 06:26:26 | INFO | train rank=0 | step=71450/100000 progress=71.5% loss=0.046188 lr=2.067766e-06 elapsed=26:30:56 avg_step=1.34s recent_step=1.22s eta=10:35:42 +2026-08-26 06:26:37 | INFO | train rank=0 | step=71460/100000 progress=71.5% loss=0.045868 lr=2.066427e-06 elapsed=26:31:07 avg_step=1.34s recent_step=1.10s eta=10:35:28 +2026-08-26 06:26:49 | INFO | train rank=0 | step=71470/100000 progress=71.5% loss=0.053688 lr=2.065088e-06 elapsed=26:31:19 avg_step=1.34s recent_step=1.25s eta=10:35:14 +2026-08-26 06:27:02 | INFO | train rank=0 | step=71480/100000 progress=71.5% loss=0.027267 lr=2.063750e-06 elapsed=26:31:32 avg_step=1.34s recent_step=1.28s eta=10:35:00 +2026-08-26 06:27:16 | INFO | train rank=0 | step=71490/100000 progress=71.5% loss=0.034248 lr=2.062412e-06 elapsed=26:31:46 avg_step=1.34s recent_step=1.40s eta=10:34:47 +2026-08-26 06:27:28 | INFO | train rank=0 | step=71500/100000 progress=71.5% loss=0.040527 lr=2.061074e-06 elapsed=26:31:58 avg_step=1.34s recent_step=1.22s eta=10:34:33 +2026-08-26 06:27:42 | INFO | train rank=0 | step=71510/100000 progress=71.5% loss=0.061066 lr=2.059736e-06 elapsed=26:32:12 avg_step=1.34s recent_step=1.36s eta=10:34:20 +2026-08-26 06:27:55 | INFO | train rank=0 | step=71520/100000 progress=71.5% loss=0.069122 lr=2.058399e-06 elapsed=26:32:25 avg_step=1.34s recent_step=1.30s eta=10:34:07 +2026-08-26 06:28:11 | INFO | train rank=0 | step=71530/100000 progress=71.5% loss=0.053223 lr=2.057062e-06 elapsed=26:32:41 avg_step=1.34s recent_step=1.60s eta=10:33:54 +2026-08-26 06:28:24 | INFO | train rank=0 | step=71540/100000 progress=71.5% loss=0.050568 lr=2.055726e-06 elapsed=26:32:54 avg_step=1.34s recent_step=1.27s eta=10:33:41 +2026-08-26 06:28:38 | INFO | train rank=0 | step=71550/100000 progress=71.5% loss=0.047913 lr=2.054389e-06 elapsed=26:33:08 avg_step=1.34s recent_step=1.46s eta=10:33:28 +2026-08-26 06:28:52 | INFO | train rank=0 | step=71560/100000 progress=71.6% loss=0.065399 lr=2.053053e-06 elapsed=26:33:22 avg_step=1.34s recent_step=1.34s eta=10:33:15 +2026-08-26 06:29:07 | INFO | train rank=0 | step=71570/100000 progress=71.6% loss=0.048508 lr=2.051718e-06 elapsed=26:33:37 avg_step=1.34s recent_step=1.53s eta=10:33:02 +2026-08-26 06:29:19 | INFO | train rank=0 | step=71580/100000 progress=71.6% loss=0.077545 lr=2.050383e-06 elapsed=26:33:49 avg_step=1.34s recent_step=1.25s eta=10:32:48 +2026-08-26 06:29:33 | INFO | train rank=0 | step=71590/100000 progress=71.6% loss=0.027634 lr=2.049048e-06 elapsed=26:34:02 avg_step=1.34s recent_step=1.31s eta=10:32:35 +2026-08-26 06:29:47 | INFO | train rank=0 | step=71600/100000 progress=71.6% loss=0.051819 lr=2.047713e-06 elapsed=26:34:17 avg_step=1.34s recent_step=1.47s eta=10:32:22 +2026-08-26 06:30:03 | INFO | train rank=0 | step=71610/100000 progress=71.6% loss=0.065857 lr=2.046379e-06 elapsed=26:34:32 avg_step=1.34s recent_step=1.53s eta=10:32:09 +2026-08-26 06:30:15 | INFO | train rank=0 | step=71620/100000 progress=71.6% loss=0.033684 lr=2.045045e-06 elapsed=26:34:45 avg_step=1.34s recent_step=1.25s eta=10:31:56 +2026-08-26 06:30:28 | INFO | train rank=0 | step=71630/100000 progress=71.6% loss=0.061310 lr=2.043711e-06 elapsed=26:34:58 avg_step=1.34s recent_step=1.28s eta=10:31:42 +2026-08-26 06:30:40 | INFO | train rank=0 | step=71640/100000 progress=71.6% loss=0.053757 lr=2.042378e-06 elapsed=26:35:10 avg_step=1.34s recent_step=1.25s eta=10:31:28 +2026-08-26 06:30:55 | INFO | train rank=0 | step=71650/100000 progress=71.7% loss=0.100037 lr=2.041045e-06 elapsed=26:35:25 avg_step=1.34s recent_step=1.49s eta=10:31:16 +2026-08-26 06:31:08 | INFO | train rank=0 | step=71660/100000 progress=71.7% loss=0.053116 lr=2.039712e-06 elapsed=26:35:37 avg_step=1.34s recent_step=1.23s eta=10:31:02 +2026-08-26 06:31:24 | INFO | train rank=0 | step=71670/100000 progress=71.7% loss=0.031189 lr=2.038380e-06 elapsed=26:35:54 avg_step=1.34s recent_step=1.69s eta=10:30:50 +2026-08-26 06:31:37 | INFO | train rank=0 | step=71680/100000 progress=71.7% loss=0.038452 lr=2.037048e-06 elapsed=26:36:07 avg_step=1.34s recent_step=1.27s eta=10:30:36 +2026-08-26 06:31:50 | INFO | train rank=0 | step=71690/100000 progress=71.7% loss=0.051636 lr=2.035716e-06 elapsed=26:36:20 avg_step=1.34s recent_step=1.29s eta=10:30:23 +2026-08-26 06:32:04 | INFO | train rank=0 | step=71700/100000 progress=71.7% loss=0.065704 lr=2.034385e-06 elapsed=26:36:34 avg_step=1.34s recent_step=1.39s eta=10:30:10 +2026-08-26 06:32:17 | INFO | train rank=0 | step=71710/100000 progress=71.7% loss=0.054993 lr=2.033053e-06 elapsed=26:36:47 avg_step=1.34s recent_step=1.30s eta=10:29:56 +2026-08-26 06:32:30 | INFO | train rank=0 | step=71720/100000 progress=71.7% loss=0.069153 lr=2.031723e-06 elapsed=26:37:00 avg_step=1.34s recent_step=1.34s eta=10:29:43 +2026-08-26 06:32:45 | INFO | train rank=0 | step=71730/100000 progress=71.7% loss=0.067780 lr=2.030392e-06 elapsed=26:37:15 avg_step=1.34s recent_step=1.44s eta=10:29:30 +2026-08-26 06:32:58 | INFO | train rank=0 | step=71740/100000 progress=71.7% loss=0.031494 lr=2.029062e-06 elapsed=26:37:28 avg_step=1.34s recent_step=1.30s eta=10:29:16 +2026-08-26 06:33:10 | INFO | train rank=0 | step=71750/100000 progress=71.8% loss=0.039764 lr=2.027732e-06 elapsed=26:37:40 avg_step=1.34s recent_step=1.27s eta=10:29:03 +2026-08-26 06:33:25 | INFO | train rank=0 | step=71760/100000 progress=71.8% loss=0.038956 lr=2.026403e-06 elapsed=26:37:55 avg_step=1.34s recent_step=1.43s eta=10:28:50 +2026-08-26 06:33:37 | INFO | train rank=0 | step=71770/100000 progress=71.8% loss=0.059387 lr=2.025074e-06 elapsed=26:38:07 avg_step=1.34s recent_step=1.24s eta=10:28:36 +2026-08-26 06:33:52 | INFO | train rank=0 | step=71780/100000 progress=71.8% loss=0.046883 lr=2.023745e-06 elapsed=26:38:22 avg_step=1.34s recent_step=1.45s eta=10:28:23 +2026-08-26 06:34:09 | INFO | train rank=0 | step=71790/100000 progress=71.8% loss=0.046356 lr=2.022417e-06 elapsed=26:38:39 avg_step=1.34s recent_step=1.77s eta=10:28:11 +2026-08-26 06:34:22 | INFO | train rank=0 | step=71800/100000 progress=71.8% loss=0.097656 lr=2.021088e-06 elapsed=26:38:52 avg_step=1.34s recent_step=1.25s eta=10:27:58 +2026-08-26 06:34:35 | INFO | train rank=0 | step=71810/100000 progress=71.8% loss=0.072937 lr=2.019761e-06 elapsed=26:39:05 avg_step=1.34s recent_step=1.35s eta=10:27:44 +2026-08-26 06:34:48 | INFO | train rank=0 | step=71820/100000 progress=71.8% loss=0.039276 lr=2.018433e-06 elapsed=26:39:18 avg_step=1.34s recent_step=1.27s eta=10:27:31 +2026-08-26 06:35:01 | INFO | train rank=0 | step=71830/100000 progress=71.8% loss=0.034401 lr=2.017106e-06 elapsed=26:39:30 avg_step=1.34s recent_step=1.25s eta=10:27:17 +2026-08-26 06:35:13 | INFO | train rank=0 | step=71840/100000 progress=71.8% loss=0.035744 lr=2.015779e-06 elapsed=26:39:43 avg_step=1.34s recent_step=1.23s eta=10:27:03 +2026-08-26 06:35:27 | INFO | train rank=0 | step=71850/100000 progress=71.9% loss=0.044327 lr=2.014453e-06 elapsed=26:39:57 avg_step=1.34s recent_step=1.40s eta=10:26:50 +2026-08-26 06:35:40 | INFO | train rank=0 | step=71860/100000 progress=71.9% loss=0.045670 lr=2.013126e-06 elapsed=26:40:10 avg_step=1.34s recent_step=1.28s eta=10:26:37 +2026-08-26 06:35:52 | INFO | train rank=0 | step=71870/100000 progress=71.9% loss=0.030502 lr=2.011801e-06 elapsed=26:40:22 avg_step=1.34s recent_step=1.25s eta=10:26:23 +2026-08-26 06:36:04 | INFO | train rank=0 | step=71880/100000 progress=71.9% loss=0.051697 lr=2.010475e-06 elapsed=26:40:34 avg_step=1.34s recent_step=1.23s eta=10:26:09 +2026-08-26 06:36:18 | INFO | train rank=0 | step=71890/100000 progress=71.9% loss=0.036240 lr=2.009150e-06 elapsed=26:40:48 avg_step=1.34s recent_step=1.34s eta=10:25:56 +2026-08-26 06:36:30 | INFO | train rank=0 | step=71900/100000 progress=71.9% loss=0.045151 lr=2.007825e-06 elapsed=26:41:00 avg_step=1.34s recent_step=1.24s eta=10:25:42 +2026-08-26 06:36:43 | INFO | train rank=0 | step=71910/100000 progress=71.9% loss=0.041931 lr=2.006500e-06 elapsed=26:41:13 avg_step=1.34s recent_step=1.32s eta=10:25:29 +2026-08-26 06:36:56 | INFO | train rank=0 | step=71920/100000 progress=71.9% loss=0.046860 lr=2.005176e-06 elapsed=26:41:26 avg_step=1.34s recent_step=1.26s eta=10:25:15 +2026-08-26 06:37:09 | INFO | train rank=0 | step=71930/100000 progress=71.9% loss=0.031845 lr=2.003852e-06 elapsed=26:41:39 avg_step=1.34s recent_step=1.34s eta=10:25:02 +2026-08-26 06:37:23 | INFO | train rank=0 | step=71940/100000 progress=71.9% loss=0.027565 lr=2.002529e-06 elapsed=26:41:53 avg_step=1.34s recent_step=1.35s eta=10:24:48 +2026-08-26 06:37:35 | INFO | train rank=0 | step=71950/100000 progress=72.0% loss=0.035690 lr=2.001206e-06 elapsed=26:42:05 avg_step=1.34s recent_step=1.26s eta=10:24:35 +2026-08-26 06:37:50 | INFO | train rank=0 | step=71960/100000 progress=72.0% loss=0.043198 lr=1.999883e-06 elapsed=26:42:20 avg_step=1.34s recent_step=1.49s eta=10:24:22 +2026-08-26 06:38:08 | INFO | train rank=0 | step=71970/100000 progress=72.0% loss=0.034653 lr=1.998560e-06 elapsed=26:42:38 avg_step=1.34s recent_step=1.76s eta=10:24:10 +2026-08-26 06:38:22 | INFO | train rank=0 | step=71980/100000 progress=72.0% loss=0.077240 lr=1.997238e-06 elapsed=26:42:52 avg_step=1.34s recent_step=1.42s eta=10:23:57 +2026-08-26 06:38:35 | INFO | train rank=0 | step=71990/100000 progress=72.0% loss=0.041023 lr=1.995916e-06 elapsed=26:43:05 avg_step=1.34s recent_step=1.25s eta=10:23:43 +2026-08-26 06:38:51 | INFO | train rank=0 | step=72000/100000 progress=72.0% loss=0.037033 lr=1.994594e-06 elapsed=26:43:21 avg_step=1.34s recent_step=1.61s eta=10:23:31 +2026-08-26 06:39:04 | INFO | train rank=0 | step=72010/100000 progress=72.0% loss=0.023010 lr=1.993273e-06 elapsed=26:43:34 avg_step=1.34s recent_step=1.31s eta=10:23:18 +2026-08-26 06:39:16 | INFO | train rank=0 | step=72020/100000 progress=72.0% loss=0.061890 lr=1.991952e-06 elapsed=26:43:46 avg_step=1.34s recent_step=1.22s eta=10:23:04 +2026-08-26 06:39:27 | INFO | train rank=0 | step=72030/100000 progress=72.0% loss=0.046677 lr=1.990631e-06 elapsed=26:43:57 avg_step=1.34s recent_step=1.12s eta=10:22:50 +2026-08-26 06:39:40 | INFO | train rank=0 | step=72040/100000 progress=72.0% loss=0.039215 lr=1.989311e-06 elapsed=26:44:10 avg_step=1.34s recent_step=1.29s eta=10:22:36 +2026-08-26 06:39:53 | INFO | train rank=0 | step=72050/100000 progress=72.0% loss=0.017273 lr=1.987991e-06 elapsed=26:44:23 avg_step=1.34s recent_step=1.24s eta=10:22:22 +2026-08-26 06:40:05 | INFO | train rank=0 | step=72060/100000 progress=72.1% loss=0.050201 lr=1.986672e-06 elapsed=26:44:35 avg_step=1.34s recent_step=1.25s eta=10:22:09 +2026-08-26 06:40:17 | INFO | train rank=0 | step=72070/100000 progress=72.1% loss=0.026993 lr=1.985352e-06 elapsed=26:44:47 avg_step=1.34s recent_step=1.16s eta=10:21:55 +2026-08-26 06:40:30 | INFO | train rank=0 | step=72080/100000 progress=72.1% loss=0.078506 lr=1.984033e-06 elapsed=26:44:59 avg_step=1.34s recent_step=1.28s eta=10:21:41 +2026-08-26 06:40:42 | INFO | train rank=0 | step=72090/100000 progress=72.1% loss=0.028973 lr=1.982715e-06 elapsed=26:45:12 avg_step=1.34s recent_step=1.25s eta=10:21:27 +2026-08-26 06:40:54 | INFO | train rank=0 | step=72100/100000 progress=72.1% loss=0.071533 lr=1.981396e-06 elapsed=26:45:23 avg_step=1.34s recent_step=1.15s eta=10:21:13 +2026-08-26 06:41:05 | INFO | train rank=0 | step=72110/100000 progress=72.1% loss=0.113220 lr=1.980078e-06 elapsed=26:45:35 avg_step=1.34s recent_step=1.16s eta=10:20:59 +2026-08-26 06:41:18 | INFO | train rank=0 | step=72120/100000 progress=72.1% loss=0.027004 lr=1.978761e-06 elapsed=26:45:48 avg_step=1.34s recent_step=1.31s eta=10:20:46 +2026-08-26 06:41:31 | INFO | train rank=0 | step=72130/100000 progress=72.1% loss=0.054962 lr=1.977443e-06 elapsed=26:46:00 avg_step=1.34s recent_step=1.23s eta=10:20:32 +2026-08-26 06:41:46 | INFO | train rank=0 | step=72140/100000 progress=72.1% loss=0.040436 lr=1.976126e-06 elapsed=26:46:16 avg_step=1.34s recent_step=1.52s eta=10:20:19 +2026-08-26 06:42:00 | INFO | train rank=0 | step=72150/100000 progress=72.2% loss=0.066330 lr=1.974810e-06 elapsed=26:46:30 avg_step=1.34s recent_step=1.41s eta=10:20:06 +2026-08-26 06:42:13 | INFO | train rank=0 | step=72160/100000 progress=72.2% loss=0.074280 lr=1.973493e-06 elapsed=26:46:43 avg_step=1.34s recent_step=1.28s eta=10:19:53 +2026-08-26 06:42:29 | INFO | train rank=0 | step=72170/100000 progress=72.2% loss=0.043854 lr=1.972178e-06 elapsed=26:46:59 avg_step=1.34s recent_step=1.60s eta=10:19:40 +2026-08-26 06:42:43 | INFO | train rank=0 | step=72180/100000 progress=72.2% loss=0.049057 lr=1.970862e-06 elapsed=26:47:13 avg_step=1.34s recent_step=1.45s eta=10:19:27 +2026-08-26 06:42:58 | INFO | train rank=0 | step=72190/100000 progress=72.2% loss=0.076050 lr=1.969547e-06 elapsed=26:47:28 avg_step=1.34s recent_step=1.49s eta=10:19:15 +2026-08-26 06:43:14 | INFO | train rank=0 | step=72200/100000 progress=72.2% loss=0.039444 lr=1.968232e-06 elapsed=26:47:44 avg_step=1.34s recent_step=1.59s eta=10:19:02 +2026-08-26 06:43:26 | INFO | train rank=0 | step=72210/100000 progress=72.2% loss=0.044388 lr=1.966917e-06 elapsed=26:47:56 avg_step=1.34s recent_step=1.21s eta=10:18:48 +2026-08-26 06:43:38 | INFO | train rank=0 | step=72220/100000 progress=72.2% loss=0.053421 lr=1.965603e-06 elapsed=26:48:08 avg_step=1.34s recent_step=1.18s eta=10:18:34 +2026-08-26 06:43:52 | INFO | train rank=0 | step=72230/100000 progress=72.2% loss=0.052261 lr=1.964289e-06 elapsed=26:48:22 avg_step=1.34s recent_step=1.46s eta=10:18:22 +2026-08-26 06:44:07 | INFO | train rank=0 | step=72240/100000 progress=72.2% loss=0.043274 lr=1.962975e-06 elapsed=26:48:37 avg_step=1.34s recent_step=1.45s eta=10:18:09 +2026-08-26 06:44:21 | INFO | train rank=0 | step=72250/100000 progress=72.2% loss=0.037628 lr=1.961662e-06 elapsed=26:48:51 avg_step=1.34s recent_step=1.40s eta=10:17:56 +2026-08-26 06:44:33 | INFO | train rank=0 | step=72260/100000 progress=72.3% loss=0.113525 lr=1.960349e-06 elapsed=26:49:03 avg_step=1.34s recent_step=1.19s eta=10:17:42 +2026-08-26 06:44:45 | INFO | train rank=0 | step=72270/100000 progress=72.3% loss=0.042557 lr=1.959036e-06 elapsed=26:49:15 avg_step=1.34s recent_step=1.17s eta=10:17:28 +2026-08-26 06:44:58 | INFO | train rank=0 | step=72280/100000 progress=72.3% loss=0.034805 lr=1.957723e-06 elapsed=26:49:28 avg_step=1.34s recent_step=1.38s eta=10:17:14 +2026-08-26 06:45:18 | INFO | train rank=0 | step=72290/100000 progress=72.3% loss=0.043739 lr=1.956411e-06 elapsed=26:49:48 avg_step=1.34s recent_step=1.95s eta=10:17:03 +2026-08-26 06:45:29 | INFO | train rank=0 | step=72300/100000 progress=72.3% loss=0.032875 lr=1.955100e-06 elapsed=26:49:59 avg_step=1.34s recent_step=1.14s eta=10:16:49 +2026-08-26 06:45:43 | INFO | train rank=0 | step=72310/100000 progress=72.3% loss=0.031586 lr=1.953788e-06 elapsed=26:50:13 avg_step=1.34s recent_step=1.36s eta=10:16:36 +2026-08-26 06:45:58 | INFO | train rank=0 | step=72320/100000 progress=72.3% loss=0.045044 lr=1.952477e-06 elapsed=26:50:28 avg_step=1.34s recent_step=1.48s eta=10:16:23 +2026-08-26 06:46:09 | INFO | train rank=0 | step=72330/100000 progress=72.3% loss=0.059891 lr=1.951167e-06 elapsed=26:50:39 avg_step=1.34s recent_step=1.17s eta=10:16:09 +2026-08-26 06:46:24 | INFO | train rank=0 | step=72340/100000 progress=72.3% loss=0.038620 lr=1.949856e-06 elapsed=26:50:54 avg_step=1.34s recent_step=1.42s eta=10:15:56 +2026-08-26 06:46:35 | INFO | train rank=0 | step=72350/100000 progress=72.4% loss=0.030556 lr=1.948546e-06 elapsed=26:51:05 avg_step=1.34s recent_step=1.18s eta=10:15:42 +2026-08-26 06:46:47 | INFO | train rank=0 | step=72360/100000 progress=72.4% loss=0.041443 lr=1.947237e-06 elapsed=26:51:17 avg_step=1.34s recent_step=1.12s eta=10:15:28 +2026-08-26 06:47:04 | INFO | train rank=0 | step=72370/100000 progress=72.4% loss=0.039864 lr=1.945927e-06 elapsed=26:51:34 avg_step=1.34s recent_step=1.72s eta=10:15:16 +2026-08-26 06:47:22 | INFO | train rank=0 | step=72380/100000 progress=72.4% loss=0.049698 lr=1.944618e-06 elapsed=26:51:52 avg_step=1.34s recent_step=1.78s eta=10:15:05 +2026-08-26 06:47:34 | INFO | train rank=0 | step=72390/100000 progress=72.4% loss=0.032059 lr=1.943310e-06 elapsed=26:52:04 avg_step=1.34s recent_step=1.26s eta=10:14:51 +2026-08-26 06:47:50 | INFO | train rank=0 | step=72400/100000 progress=72.4% loss=0.048950 lr=1.942001e-06 elapsed=26:52:19 avg_step=1.34s recent_step=1.53s eta=10:14:38 +2026-08-26 06:48:02 | INFO | train rank=0 | step=72410/100000 progress=72.4% loss=0.044250 lr=1.940693e-06 elapsed=26:52:32 avg_step=1.34s recent_step=1.29s eta=10:14:25 +2026-08-26 06:48:15 | INFO | train rank=0 | step=72420/100000 progress=72.4% loss=0.068848 lr=1.939386e-06 elapsed=26:52:45 avg_step=1.34s recent_step=1.27s eta=10:14:11 +2026-08-26 06:48:28 | INFO | train rank=0 | step=72430/100000 progress=72.4% loss=0.045013 lr=1.938078e-06 elapsed=26:52:58 avg_step=1.34s recent_step=1.25s eta=10:13:57 +2026-08-26 06:48:40 | INFO | train rank=0 | step=72440/100000 progress=72.4% loss=0.027580 lr=1.936771e-06 elapsed=26:53:10 avg_step=1.34s recent_step=1.23s eta=10:13:44 +2026-08-26 06:48:54 | INFO | train rank=0 | step=72450/100000 progress=72.5% loss=0.046921 lr=1.935465e-06 elapsed=26:53:24 avg_step=1.34s recent_step=1.38s eta=10:13:30 +2026-08-26 06:49:06 | INFO | train rank=0 | step=72460/100000 progress=72.5% loss=0.039116 lr=1.934158e-06 elapsed=26:53:36 avg_step=1.34s recent_step=1.19s eta=10:13:17 +2026-08-26 06:49:20 | INFO | train rank=0 | step=72470/100000 progress=72.5% loss=0.036041 lr=1.932852e-06 elapsed=26:53:50 avg_step=1.34s recent_step=1.45s eta=10:13:04 +2026-08-26 06:49:35 | INFO | train rank=0 | step=72480/100000 progress=72.5% loss=0.039520 lr=1.931547e-06 elapsed=26:54:05 avg_step=1.34s recent_step=1.51s eta=10:12:51 +2026-08-26 06:49:50 | INFO | train rank=0 | step=72490/100000 progress=72.5% loss=0.044876 lr=1.930241e-06 elapsed=26:54:20 avg_step=1.34s recent_step=1.48s eta=10:12:38 +2026-08-26 06:50:06 | INFO | train rank=0 | step=72500/100000 progress=72.5% loss=0.050003 lr=1.928936e-06 elapsed=26:54:36 avg_step=1.34s recent_step=1.62s eta=10:12:26 +2026-08-26 06:50:18 | INFO | train rank=0 | step=72510/100000 progress=72.5% loss=0.052536 lr=1.927632e-06 elapsed=26:54:48 avg_step=1.34s recent_step=1.20s eta=10:12:12 +2026-08-26 06:50:31 | INFO | train rank=0 | step=72520/100000 progress=72.5% loss=0.045166 lr=1.926327e-06 elapsed=26:55:01 avg_step=1.34s recent_step=1.27s eta=10:11:58 +2026-08-26 06:50:45 | INFO | train rank=0 | step=72530/100000 progress=72.5% loss=0.049057 lr=1.925023e-06 elapsed=26:55:15 avg_step=1.34s recent_step=1.44s eta=10:11:45 +2026-08-26 06:50:57 | INFO | train rank=0 | step=72540/100000 progress=72.5% loss=0.053604 lr=1.923720e-06 elapsed=26:55:27 avg_step=1.34s recent_step=1.21s eta=10:11:32 +2026-08-26 06:51:10 | INFO | train rank=0 | step=72550/100000 progress=72.5% loss=0.063110 lr=1.922417e-06 elapsed=26:55:40 avg_step=1.34s recent_step=1.29s eta=10:11:18 +2026-08-26 06:51:23 | INFO | train rank=0 | step=72560/100000 progress=72.6% loss=0.082733 lr=1.921114e-06 elapsed=26:55:53 avg_step=1.34s recent_step=1.26s eta=10:11:04 +2026-08-26 06:51:38 | INFO | train rank=0 | step=72570/100000 progress=72.6% loss=0.050110 lr=1.919811e-06 elapsed=26:56:07 avg_step=1.34s recent_step=1.47s eta=10:10:51 +2026-08-26 06:51:52 | INFO | train rank=0 | step=72580/100000 progress=72.6% loss=0.036514 lr=1.918509e-06 elapsed=26:56:22 avg_step=1.34s recent_step=1.49s eta=10:10:39 +2026-08-26 06:52:05 | INFO | train rank=0 | step=72590/100000 progress=72.6% loss=0.046829 lr=1.917207e-06 elapsed=26:56:34 avg_step=1.34s recent_step=1.21s eta=10:10:25 +2026-08-26 06:52:17 | INFO | train rank=0 | step=72600/100000 progress=72.6% loss=0.060593 lr=1.915905e-06 elapsed=26:56:47 avg_step=1.34s recent_step=1.26s eta=10:10:11 +2026-08-26 06:52:35 | INFO | train rank=0 | step=72610/100000 progress=72.6% loss=0.038696 lr=1.914604e-06 elapsed=26:57:05 avg_step=1.34s recent_step=1.82s eta=10:10:00 +2026-08-26 06:52:49 | INFO | train rank=0 | step=72620/100000 progress=72.6% loss=0.058243 lr=1.913303e-06 elapsed=26:57:18 avg_step=1.34s recent_step=1.32s eta=10:09:46 +2026-08-26 06:53:01 | INFO | train rank=0 | step=72630/100000 progress=72.6% loss=0.044312 lr=1.912002e-06 elapsed=26:57:31 avg_step=1.34s recent_step=1.26s eta=10:09:33 +2026-08-26 06:53:14 | INFO | train rank=0 | step=72640/100000 progress=72.6% loss=0.059418 lr=1.910702e-06 elapsed=26:57:44 avg_step=1.34s recent_step=1.24s eta=10:09:19 +2026-08-26 06:53:28 | INFO | train rank=0 | step=72650/100000 progress=72.7% loss=0.044212 lr=1.909402e-06 elapsed=26:57:58 avg_step=1.34s recent_step=1.42s eta=10:09:06 +2026-08-26 06:53:41 | INFO | train rank=0 | step=72660/100000 progress=72.7% loss=0.047562 lr=1.908102e-06 elapsed=26:58:10 avg_step=1.34s recent_step=1.27s eta=10:08:52 +2026-08-26 06:53:54 | INFO | train rank=0 | step=72670/100000 progress=72.7% loss=0.055313 lr=1.906803e-06 elapsed=26:58:24 avg_step=1.34s recent_step=1.35s eta=10:08:39 +2026-08-26 06:54:06 | INFO | train rank=0 | step=72680/100000 progress=72.7% loss=0.059265 lr=1.905504e-06 elapsed=26:58:36 avg_step=1.34s recent_step=1.23s eta=10:08:25 +2026-08-26 06:54:19 | INFO | train rank=0 | step=72690/100000 progress=72.7% loss=0.094208 lr=1.904206e-06 elapsed=26:58:49 avg_step=1.34s recent_step=1.30s eta=10:08:12 +2026-08-26 06:54:35 | INFO | train rank=0 | step=72700/100000 progress=72.7% loss=0.045303 lr=1.902907e-06 elapsed=26:59:05 avg_step=1.34s recent_step=1.60s eta=10:07:59 +2026-08-26 06:54:47 | INFO | train rank=0 | step=72710/100000 progress=72.7% loss=0.044907 lr=1.901609e-06 elapsed=26:59:17 avg_step=1.34s recent_step=1.16s eta=10:07:45 +2026-08-26 06:54:58 | INFO | train rank=0 | step=72720/100000 progress=72.7% loss=0.042938 lr=1.900312e-06 elapsed=26:59:28 avg_step=1.34s recent_step=1.16s eta=10:07:31 +2026-08-26 06:55:11 | INFO | train rank=0 | step=72730/100000 progress=72.7% loss=0.045563 lr=1.899015e-06 elapsed=26:59:41 avg_step=1.34s recent_step=1.24s eta=10:07:17 +2026-08-26 06:55:24 | INFO | train rank=0 | step=72740/100000 progress=72.7% loss=0.047394 lr=1.897718e-06 elapsed=26:59:54 avg_step=1.34s recent_step=1.28s eta=10:07:04 +2026-08-26 06:55:37 | INFO | train rank=0 | step=72750/100000 progress=72.8% loss=0.025780 lr=1.896421e-06 elapsed=27:00:07 avg_step=1.34s recent_step=1.37s eta=10:06:51 +2026-08-26 06:55:49 | INFO | train rank=0 | step=72760/100000 progress=72.8% loss=0.056168 lr=1.895125e-06 elapsed=27:00:19 avg_step=1.34s recent_step=1.18s eta=10:06:37 +2026-08-26 06:56:05 | INFO | train rank=0 | step=72770/100000 progress=72.8% loss=0.039169 lr=1.893829e-06 elapsed=27:00:35 avg_step=1.34s recent_step=1.60s eta=10:06:24 +2026-08-26 06:56:17 | INFO | train rank=0 | step=72780/100000 progress=72.8% loss=0.042084 lr=1.892534e-06 elapsed=27:00:47 avg_step=1.34s recent_step=1.15s eta=10:06:10 +2026-08-26 06:56:30 | INFO | train rank=0 | step=72790/100000 progress=72.8% loss=0.074768 lr=1.891238e-06 elapsed=27:01:00 avg_step=1.34s recent_step=1.32s eta=10:05:57 +2026-08-26 06:56:42 | INFO | train rank=0 | step=72800/100000 progress=72.8% loss=0.043854 lr=1.889944e-06 elapsed=27:01:12 avg_step=1.34s recent_step=1.18s eta=10:05:43 +2026-08-26 06:56:57 | INFO | train rank=0 | step=72810/100000 progress=72.8% loss=0.075134 lr=1.888649e-06 elapsed=27:01:27 avg_step=1.34s recent_step=1.55s eta=10:05:30 +2026-08-26 06:57:13 | INFO | train rank=0 | step=72820/100000 progress=72.8% loss=0.049698 lr=1.887355e-06 elapsed=27:01:43 avg_step=1.34s recent_step=1.57s eta=10:05:18 +2026-08-26 06:57:24 | INFO | train rank=0 | step=72830/100000 progress=72.8% loss=0.053009 lr=1.886061e-06 elapsed=27:01:54 avg_step=1.34s recent_step=1.14s eta=10:05:04 +2026-08-26 06:57:36 | INFO | train rank=0 | step=72840/100000 progress=72.8% loss=0.033524 lr=1.884768e-06 elapsed=27:02:06 avg_step=1.34s recent_step=1.18s eta=10:04:50 +2026-08-26 06:57:49 | INFO | train rank=0 | step=72850/100000 progress=72.9% loss=0.065704 lr=1.883474e-06 elapsed=27:02:19 avg_step=1.34s recent_step=1.29s eta=10:04:36 +2026-08-26 06:58:00 | INFO | train rank=0 | step=72860/100000 progress=72.9% loss=0.035217 lr=1.882182e-06 elapsed=27:02:30 avg_step=1.34s recent_step=1.16s eta=10:04:22 +2026-08-26 06:58:16 | INFO | train rank=0 | step=72870/100000 progress=72.9% loss=0.033432 lr=1.880889e-06 elapsed=27:02:46 avg_step=1.34s recent_step=1.55s eta=10:04:10 +2026-08-26 06:58:30 | INFO | train rank=0 | step=72880/100000 progress=72.9% loss=0.039948 lr=1.879597e-06 elapsed=27:03:00 avg_step=1.34s recent_step=1.41s eta=10:03:57 +2026-08-26 06:58:42 | INFO | train rank=0 | step=72890/100000 progress=72.9% loss=0.049179 lr=1.878305e-06 elapsed=27:03:12 avg_step=1.34s recent_step=1.17s eta=10:03:43 +2026-08-26 06:58:58 | INFO | train rank=0 | step=72900/100000 progress=72.9% loss=0.041458 lr=1.877014e-06 elapsed=27:03:28 avg_step=1.34s recent_step=1.60s eta=10:03:30 +2026-08-26 06:59:13 | INFO | train rank=0 | step=72910/100000 progress=72.9% loss=0.044624 lr=1.875723e-06 elapsed=27:03:43 avg_step=1.34s recent_step=1.55s eta=10:03:18 +2026-08-26 06:59:25 | INFO | train rank=0 | step=72920/100000 progress=72.9% loss=0.034760 lr=1.874432e-06 elapsed=27:03:55 avg_step=1.34s recent_step=1.21s eta=10:03:04 +2026-08-26 06:59:38 | INFO | train rank=0 | step=72930/100000 progress=72.9% loss=0.054810 lr=1.873142e-06 elapsed=27:04:08 avg_step=1.34s recent_step=1.29s eta=10:02:50 +2026-08-26 06:59:54 | INFO | train rank=0 | step=72940/100000 progress=72.9% loss=0.043610 lr=1.871851e-06 elapsed=27:04:23 avg_step=1.34s recent_step=1.52s eta=10:02:38 +2026-08-26 07:00:05 | INFO | train rank=0 | step=72950/100000 progress=73.0% loss=0.051239 lr=1.870562e-06 elapsed=27:04:35 avg_step=1.34s recent_step=1.16s eta=10:02:24 +2026-08-26 07:00:17 | INFO | train rank=0 | step=72960/100000 progress=73.0% loss=0.050568 lr=1.869272e-06 elapsed=27:04:47 avg_step=1.34s recent_step=1.18s eta=10:02:10 +2026-08-26 07:00:30 | INFO | train rank=0 | step=72970/100000 progress=73.0% loss=0.039291 lr=1.867983e-06 elapsed=27:05:00 avg_step=1.34s recent_step=1.29s eta=10:01:56 +2026-08-26 07:00:47 | INFO | train rank=0 | step=72980/100000 progress=73.0% loss=0.065460 lr=1.866695e-06 elapsed=27:05:17 avg_step=1.34s recent_step=1.72s eta=10:01:44 +2026-08-26 07:00:58 | INFO | train rank=0 | step=72990/100000 progress=73.0% loss=0.053642 lr=1.865406e-06 elapsed=27:05:28 avg_step=1.34s recent_step=1.13s eta=10:01:30 +2026-08-26 07:01:12 | INFO | train rank=0 | step=73000/100000 progress=73.0% loss=0.039703 lr=1.864118e-06 elapsed=27:05:42 avg_step=1.34s recent_step=1.40s eta=10:01:17 +2026-08-26 07:01:24 | INFO | train rank=0 | step=73010/100000 progress=73.0% loss=0.041054 lr=1.862831e-06 elapsed=27:05:54 avg_step=1.34s recent_step=1.15s eta=10:01:03 +2026-08-26 07:01:35 | INFO | train rank=0 | step=73020/100000 progress=73.0% loss=0.034073 lr=1.861543e-06 elapsed=27:06:05 avg_step=1.34s recent_step=1.16s eta=10:00:49 +2026-08-26 07:01:50 | INFO | train rank=0 | step=73030/100000 progress=73.0% loss=0.052948 lr=1.860256e-06 elapsed=27:06:20 avg_step=1.34s recent_step=1.49s eta=10:00:36 +2026-08-26 07:02:02 | INFO | train rank=0 | step=73040/100000 progress=73.0% loss=0.048660 lr=1.858970e-06 elapsed=27:06:32 avg_step=1.34s recent_step=1.21s eta=10:00:22 +2026-08-26 07:02:20 | INFO | train rank=0 | step=73050/100000 progress=73.0% loss=0.052933 lr=1.857683e-06 elapsed=27:06:49 avg_step=1.34s recent_step=1.72s eta=10:00:10 +2026-08-26 07:02:33 | INFO | train rank=0 | step=73060/100000 progress=73.1% loss=0.034363 lr=1.856397e-06 elapsed=27:07:03 avg_step=1.34s recent_step=1.39s eta=09:59:57 +2026-08-26 07:02:45 | INFO | train rank=0 | step=73070/100000 progress=73.1% loss=0.072266 lr=1.855112e-06 elapsed=27:07:15 avg_step=1.34s recent_step=1.16s eta=09:59:43 +2026-08-26 07:02:59 | INFO | train rank=0 | step=73080/100000 progress=73.1% loss=0.041183 lr=1.853826e-06 elapsed=27:07:29 avg_step=1.34s recent_step=1.37s eta=09:59:30 +2026-08-26 07:03:13 | INFO | train rank=0 | step=73090/100000 progress=73.1% loss=0.056992 lr=1.852541e-06 elapsed=27:07:43 avg_step=1.34s recent_step=1.43s eta=09:59:17 +2026-08-26 07:03:26 | INFO | train rank=0 | step=73100/100000 progress=73.1% loss=0.043945 lr=1.851257e-06 elapsed=27:07:55 avg_step=1.34s recent_step=1.25s eta=09:59:03 +2026-08-26 07:03:39 | INFO | train rank=0 | step=73110/100000 progress=73.1% loss=0.062683 lr=1.849973e-06 elapsed=27:08:09 avg_step=1.34s recent_step=1.37s eta=09:58:50 +2026-08-26 07:03:52 | INFO | train rank=0 | step=73120/100000 progress=73.1% loss=0.033039 lr=1.848689e-06 elapsed=27:08:22 avg_step=1.34s recent_step=1.26s eta=09:58:36 +2026-08-26 07:04:09 | INFO | train rank=0 | step=73130/100000 progress=73.1% loss=0.054489 lr=1.847405e-06 elapsed=27:08:39 avg_step=1.34s recent_step=1.75s eta=09:58:24 +2026-08-26 07:04:22 | INFO | train rank=0 | step=73140/100000 progress=73.1% loss=0.069855 lr=1.846122e-06 elapsed=27:08:52 avg_step=1.34s recent_step=1.24s eta=09:58:11 +2026-08-26 07:04:34 | INFO | train rank=0 | step=73150/100000 progress=73.2% loss=0.045044 lr=1.844839e-06 elapsed=27:09:03 avg_step=1.34s recent_step=1.17s eta=09:57:57 +2026-08-26 07:04:45 | INFO | train rank=0 | step=73160/100000 progress=73.2% loss=0.037476 lr=1.843557e-06 elapsed=27:09:15 avg_step=1.34s recent_step=1.12s eta=09:57:43 +2026-08-26 07:04:58 | INFO | train rank=0 | step=73170/100000 progress=73.2% loss=0.028267 lr=1.842274e-06 elapsed=27:09:28 avg_step=1.34s recent_step=1.32s eta=09:57:29 +2026-08-26 07:05:11 | INFO | train rank=0 | step=73180/100000 progress=73.2% loss=0.049347 lr=1.840993e-06 elapsed=27:09:41 avg_step=1.34s recent_step=1.32s eta=09:57:16 +2026-08-26 07:05:23 | INFO | train rank=0 | step=73190/100000 progress=73.2% loss=0.028023 lr=1.839711e-06 elapsed=27:09:53 avg_step=1.34s recent_step=1.19s eta=09:57:02 +2026-08-26 07:05:40 | INFO | train rank=0 | step=73200/100000 progress=73.2% loss=0.045319 lr=1.838430e-06 elapsed=27:10:10 avg_step=1.34s recent_step=1.72s eta=09:56:50 +2026-08-26 07:05:55 | INFO | train rank=0 | step=73210/100000 progress=73.2% loss=0.048065 lr=1.837149e-06 elapsed=27:10:25 avg_step=1.34s recent_step=1.52s eta=09:56:37 +2026-08-26 07:06:09 | INFO | train rank=0 | step=73220/100000 progress=73.2% loss=0.061646 lr=1.835869e-06 elapsed=27:10:39 avg_step=1.34s recent_step=1.36s eta=09:56:24 +2026-08-26 07:06:22 | INFO | train rank=0 | step=73230/100000 progress=73.2% loss=0.046402 lr=1.834589e-06 elapsed=27:10:52 avg_step=1.34s recent_step=1.30s eta=09:56:10 +2026-08-26 07:06:40 | INFO | train rank=0 | step=73240/100000 progress=73.2% loss=0.053131 lr=1.833309e-06 elapsed=27:11:10 avg_step=1.34s recent_step=1.76s eta=09:55:59 +2026-08-26 07:06:53 | INFO | train rank=0 | step=73250/100000 progress=73.2% loss=0.041336 lr=1.832029e-06 elapsed=27:11:23 avg_step=1.34s recent_step=1.36s eta=09:55:45 +2026-08-26 07:07:08 | INFO | train rank=0 | step=73260/100000 progress=73.3% loss=0.045639 lr=1.830750e-06 elapsed=27:11:38 avg_step=1.34s recent_step=1.52s eta=09:55:33 +2026-08-26 07:07:23 | INFO | train rank=0 | step=73270/100000 progress=73.3% loss=0.051117 lr=1.829472e-06 elapsed=27:11:53 avg_step=1.34s recent_step=1.44s eta=09:55:20 +2026-08-26 07:07:34 | INFO | train rank=0 | step=73280/100000 progress=73.3% loss=0.058655 lr=1.828193e-06 elapsed=27:12:04 avg_step=1.34s recent_step=1.12s eta=09:55:06 +2026-08-26 07:07:48 | INFO | train rank=0 | step=73290/100000 progress=73.3% loss=0.059662 lr=1.826915e-06 elapsed=27:12:18 avg_step=1.34s recent_step=1.43s eta=09:54:53 +2026-08-26 07:08:00 | INFO | train rank=0 | step=73300/100000 progress=73.3% loss=0.050171 lr=1.825638e-06 elapsed=27:12:30 avg_step=1.34s recent_step=1.18s eta=09:54:39 +2026-08-26 07:08:11 | INFO | train rank=0 | step=73310/100000 progress=73.3% loss=0.031647 lr=1.824360e-06 elapsed=27:12:41 avg_step=1.34s recent_step=1.13s eta=09:54:25 +2026-08-26 07:08:23 | INFO | train rank=0 | step=73320/100000 progress=73.3% loss=0.039894 lr=1.823083e-06 elapsed=27:12:53 avg_step=1.34s recent_step=1.16s eta=09:54:10 +2026-08-26 07:08:35 | INFO | train rank=0 | step=73330/100000 progress=73.3% loss=0.035767 lr=1.821807e-06 elapsed=27:13:05 avg_step=1.34s recent_step=1.18s eta=09:53:57 +2026-08-26 07:08:46 | INFO | train rank=0 | step=73340/100000 progress=73.3% loss=0.034302 lr=1.820530e-06 elapsed=27:13:16 avg_step=1.34s recent_step=1.12s eta=09:53:42 +2026-08-26 07:08:58 | INFO | train rank=0 | step=73350/100000 progress=73.4% loss=0.062225 lr=1.819255e-06 elapsed=27:13:28 avg_step=1.34s recent_step=1.19s eta=09:53:29 +2026-08-26 07:09:14 | INFO | train rank=0 | step=73360/100000 progress=73.4% loss=0.044083 lr=1.817979e-06 elapsed=27:13:44 avg_step=1.34s recent_step=1.60s eta=09:53:16 +2026-08-26 07:09:25 | INFO | train rank=0 | step=73370/100000 progress=73.4% loss=0.073669 lr=1.816704e-06 elapsed=27:13:55 avg_step=1.34s recent_step=1.15s eta=09:53:02 +2026-08-26 07:09:40 | INFO | train rank=0 | step=73380/100000 progress=73.4% loss=0.020866 lr=1.815429e-06 elapsed=27:14:09 avg_step=1.34s recent_step=1.41s eta=09:52:49 +2026-08-26 07:09:55 | INFO | train rank=0 | step=73390/100000 progress=73.4% loss=0.042755 lr=1.814154e-06 elapsed=27:14:25 avg_step=1.34s recent_step=1.55s eta=09:52:36 +2026-08-26 07:10:07 | INFO | train rank=0 | step=73400/100000 progress=73.4% loss=0.079498 lr=1.812880e-06 elapsed=27:14:36 avg_step=1.34s recent_step=1.14s eta=09:52:22 +2026-08-26 07:10:23 | INFO | train rank=0 | step=73410/100000 progress=73.4% loss=0.044342 lr=1.811606e-06 elapsed=27:14:53 avg_step=1.34s recent_step=1.70s eta=09:52:10 +2026-08-26 07:10:40 | INFO | train rank=0 | step=73420/100000 progress=73.4% loss=0.047455 lr=1.810333e-06 elapsed=27:15:10 avg_step=1.34s recent_step=1.64s eta=09:51:58 +2026-08-26 07:10:55 | INFO | train rank=0 | step=73430/100000 progress=73.4% loss=0.025517 lr=1.809060e-06 elapsed=27:15:25 avg_step=1.34s recent_step=1.48s eta=09:51:45 +2026-08-26 07:11:07 | INFO | train rank=0 | step=73440/100000 progress=73.4% loss=0.062073 lr=1.807787e-06 elapsed=27:15:37 avg_step=1.34s recent_step=1.22s eta=09:51:31 +2026-08-26 07:11:22 | INFO | train rank=0 | step=73450/100000 progress=73.5% loss=0.060089 lr=1.806514e-06 elapsed=27:15:52 avg_step=1.34s recent_step=1.55s eta=09:51:19 +2026-08-26 07:11:34 | INFO | train rank=0 | step=73460/100000 progress=73.5% loss=0.059174 lr=1.805242e-06 elapsed=27:16:04 avg_step=1.34s recent_step=1.19s eta=09:51:05 +2026-08-26 07:11:46 | INFO | train rank=0 | step=73470/100000 progress=73.5% loss=0.046005 lr=1.803970e-06 elapsed=27:16:16 avg_step=1.34s recent_step=1.16s eta=09:50:51 +2026-08-26 07:11:58 | INFO | train rank=0 | step=73480/100000 progress=73.5% loss=0.057343 lr=1.802699e-06 elapsed=27:16:28 avg_step=1.34s recent_step=1.22s eta=09:50:37 +2026-08-26 07:12:12 | INFO | train rank=0 | step=73490/100000 progress=73.5% loss=0.076828 lr=1.801428e-06 elapsed=27:16:42 avg_step=1.34s recent_step=1.38s eta=09:50:24 +2026-08-26 07:12:23 | INFO | train rank=0 | step=73500/100000 progress=73.5% loss=0.118530 lr=1.800157e-06 elapsed=27:16:53 avg_step=1.34s recent_step=1.14s eta=09:50:10 +2026-08-26 07:12:35 | INFO | train rank=0 | step=73510/100000 progress=73.5% loss=0.059814 lr=1.798887e-06 elapsed=27:17:05 avg_step=1.34s recent_step=1.18s eta=09:49:56 +2026-08-26 07:12:46 | INFO | train rank=0 | step=73520/100000 progress=73.5% loss=0.038406 lr=1.797617e-06 elapsed=27:17:16 avg_step=1.34s recent_step=1.12s eta=09:49:42 +2026-08-26 07:13:05 | INFO | train rank=0 | step=73530/100000 progress=73.5% loss=0.044003 lr=1.796347e-06 elapsed=27:17:35 avg_step=1.34s recent_step=1.91s eta=09:49:31 +2026-08-26 07:13:17 | INFO | train rank=0 | step=73540/100000 progress=73.5% loss=0.060989 lr=1.795078e-06 elapsed=27:17:47 avg_step=1.34s recent_step=1.15s eta=09:49:16 +2026-08-26 07:13:31 | INFO | train rank=0 | step=73550/100000 progress=73.6% loss=0.052216 lr=1.793809e-06 elapsed=27:18:00 avg_step=1.34s recent_step=1.37s eta=09:49:03 +2026-08-26 07:13:45 | INFO | train rank=0 | step=73560/100000 progress=73.6% loss=0.040161 lr=1.792540e-06 elapsed=27:18:15 avg_step=1.34s recent_step=1.43s eta=09:48:50 +2026-08-26 07:14:01 | INFO | train rank=0 | step=73570/100000 progress=73.6% loss=0.041611 lr=1.791272e-06 elapsed=27:18:31 avg_step=1.34s recent_step=1.61s eta=09:48:38 +2026-08-26 07:14:12 | INFO | train rank=0 | step=73580/100000 progress=73.6% loss=0.055466 lr=1.790004e-06 elapsed=27:18:42 avg_step=1.34s recent_step=1.12s eta=09:48:24 +2026-08-26 07:14:24 | INFO | train rank=0 | step=73590/100000 progress=73.6% loss=0.038040 lr=1.788737e-06 elapsed=27:18:54 avg_step=1.34s recent_step=1.16s eta=09:48:10 +2026-08-26 07:14:40 | INFO | train rank=0 | step=73600/100000 progress=73.6% loss=0.052277 lr=1.787470e-06 elapsed=27:19:09 avg_step=1.34s recent_step=1.59s eta=09:47:57 +2026-08-26 07:14:51 | INFO | train rank=0 | step=73610/100000 progress=73.6% loss=0.041733 lr=1.786203e-06 elapsed=27:19:21 avg_step=1.34s recent_step=1.13s eta=09:47:43 +2026-08-26 07:15:05 | INFO | train rank=0 | step=73620/100000 progress=73.6% loss=0.036308 lr=1.784936e-06 elapsed=27:19:35 avg_step=1.34s recent_step=1.43s eta=09:47:30 +2026-08-26 07:15:17 | INFO | train rank=0 | step=73630/100000 progress=73.6% loss=0.072235 lr=1.783670e-06 elapsed=27:19:47 avg_step=1.34s recent_step=1.22s eta=09:47:16 +2026-08-26 07:15:33 | INFO | train rank=0 | step=73640/100000 progress=73.6% loss=0.038605 lr=1.782404e-06 elapsed=27:20:03 avg_step=1.34s recent_step=1.55s eta=09:47:04 +2026-08-26 07:15:51 | INFO | train rank=0 | step=73650/100000 progress=73.7% loss=0.051323 lr=1.781139e-06 elapsed=27:20:21 avg_step=1.34s recent_step=1.81s eta=09:46:52 +2026-08-26 07:16:05 | INFO | train rank=0 | step=73660/100000 progress=73.7% loss=0.046936 lr=1.779874e-06 elapsed=27:20:34 avg_step=1.34s recent_step=1.35s eta=09:46:39 +2026-08-26 07:16:19 | INFO | train rank=0 | step=73670/100000 progress=73.7% loss=0.050598 lr=1.778609e-06 elapsed=27:20:49 avg_step=1.34s recent_step=1.47s eta=09:46:26 +2026-08-26 07:16:31 | INFO | train rank=0 | step=73680/100000 progress=73.7% loss=0.042648 lr=1.777345e-06 elapsed=27:21:01 avg_step=1.34s recent_step=1.15s eta=09:46:12 +2026-08-26 07:16:45 | INFO | train rank=0 | step=73690/100000 progress=73.7% loss=0.061539 lr=1.776081e-06 elapsed=27:21:15 avg_step=1.34s recent_step=1.43s eta=09:45:59 +2026-08-26 07:16:59 | INFO | train rank=0 | step=73700/100000 progress=73.7% loss=0.059334 lr=1.774817e-06 elapsed=27:21:29 avg_step=1.34s recent_step=1.37s eta=09:45:46 +2026-08-26 07:17:13 | INFO | train rank=0 | step=73710/100000 progress=73.7% loss=0.082581 lr=1.773554e-06 elapsed=27:21:43 avg_step=1.34s recent_step=1.39s eta=09:45:32 +2026-08-26 07:17:24 | INFO | train rank=0 | step=73720/100000 progress=73.7% loss=0.050018 lr=1.772291e-06 elapsed=27:21:54 avg_step=1.34s recent_step=1.15s eta=09:45:18 +2026-08-26 07:17:37 | INFO | train rank=0 | step=73730/100000 progress=73.7% loss=0.052719 lr=1.771028e-06 elapsed=27:22:07 avg_step=1.34s recent_step=1.32s eta=09:45:05 +2026-08-26 07:17:49 | INFO | train rank=0 | step=73740/100000 progress=73.7% loss=0.038361 lr=1.769766e-06 elapsed=27:22:19 avg_step=1.34s recent_step=1.15s eta=09:44:51 +2026-08-26 07:18:01 | INFO | train rank=0 | step=73750/100000 progress=73.8% loss=0.080353 lr=1.768504e-06 elapsed=27:22:30 avg_step=1.34s recent_step=1.17s eta=09:44:37 +2026-08-26 07:18:13 | INFO | train rank=0 | step=73760/100000 progress=73.8% loss=0.027496 lr=1.767242e-06 elapsed=27:22:42 avg_step=1.34s recent_step=1.21s eta=09:44:23 +2026-08-26 07:18:31 | INFO | train rank=0 | step=73770/100000 progress=73.8% loss=0.054520 lr=1.765981e-06 elapsed=27:23:01 avg_step=1.34s recent_step=1.89s eta=09:44:12 +2026-08-26 07:18:43 | INFO | train rank=0 | step=73780/100000 progress=73.8% loss=0.067810 lr=1.764720e-06 elapsed=27:23:13 avg_step=1.34s recent_step=1.13s eta=09:43:58 +2026-08-26 07:18:55 | INFO | train rank=0 | step=73790/100000 progress=73.8% loss=0.024086 lr=1.763460e-06 elapsed=27:23:24 avg_step=1.34s recent_step=1.18s eta=09:43:44 +2026-08-26 07:19:06 | INFO | train rank=0 | step=73800/100000 progress=73.8% loss=0.049271 lr=1.762200e-06 elapsed=27:23:36 avg_step=1.34s recent_step=1.13s eta=09:43:30 +2026-08-26 07:19:24 | INFO | train rank=0 | step=73810/100000 progress=73.8% loss=0.038300 lr=1.760940e-06 elapsed=27:23:54 avg_step=1.34s recent_step=1.83s eta=09:43:18 +2026-08-26 07:19:40 | INFO | train rank=0 | step=73820/100000 progress=73.8% loss=0.034622 lr=1.759680e-06 elapsed=27:24:10 avg_step=1.34s recent_step=1.58s eta=09:43:05 +2026-08-26 07:19:51 | INFO | train rank=0 | step=73830/100000 progress=73.8% loss=0.048538 lr=1.758421e-06 elapsed=27:24:21 avg_step=1.34s recent_step=1.13s eta=09:42:51 +2026-08-26 07:20:04 | INFO | train rank=0 | step=73840/100000 progress=73.8% loss=0.036148 lr=1.757163e-06 elapsed=27:24:33 avg_step=1.34s recent_step=1.23s eta=09:42:38 +2026-08-26 07:20:15 | INFO | train rank=0 | step=73850/100000 progress=73.9% loss=0.051453 lr=1.755904e-06 elapsed=27:24:45 avg_step=1.34s recent_step=1.15s eta=09:42:24 +2026-08-26 07:20:27 | INFO | train rank=0 | step=73860/100000 progress=73.9% loss=0.072021 lr=1.754646e-06 elapsed=27:24:57 avg_step=1.34s recent_step=1.17s eta=09:42:10 +2026-08-26 07:20:41 | INFO | train rank=0 | step=73870/100000 progress=73.9% loss=0.043541 lr=1.753388e-06 elapsed=27:25:11 avg_step=1.34s recent_step=1.45s eta=09:41:57 +2026-08-26 07:20:53 | INFO | train rank=0 | step=73880/100000 progress=73.9% loss=0.027435 lr=1.752131e-06 elapsed=27:25:23 avg_step=1.34s recent_step=1.16s eta=09:41:43 +2026-08-26 07:21:07 | INFO | train rank=0 | step=73890/100000 progress=73.9% loss=0.080475 lr=1.750874e-06 elapsed=27:25:37 avg_step=1.34s recent_step=1.41s eta=09:41:30 +2026-08-26 07:21:18 | INFO | train rank=0 | step=73900/100000 progress=73.9% loss=0.044708 lr=1.749618e-06 elapsed=27:25:48 avg_step=1.34s recent_step=1.15s eta=09:41:16 +2026-08-26 07:21:31 | INFO | train rank=0 | step=73910/100000 progress=73.9% loss=0.048996 lr=1.748361e-06 elapsed=27:26:01 avg_step=1.34s recent_step=1.27s eta=09:41:02 +2026-08-26 07:21:46 | INFO | train rank=0 | step=73920/100000 progress=73.9% loss=0.089355 lr=1.747105e-06 elapsed=27:26:16 avg_step=1.34s recent_step=1.52s eta=09:40:49 +2026-08-26 07:22:02 | INFO | train rank=0 | step=73930/100000 progress=73.9% loss=0.047691 lr=1.745850e-06 elapsed=27:26:32 avg_step=1.34s recent_step=1.56s eta=09:40:37 +2026-08-26 07:22:13 | INFO | train rank=0 | step=73940/100000 progress=73.9% loss=0.061111 lr=1.744595e-06 elapsed=27:26:43 avg_step=1.34s recent_step=1.14s eta=09:40:23 +2026-08-26 07:22:31 | INFO | train rank=0 | step=73950/100000 progress=74.0% loss=0.059723 lr=1.743340e-06 elapsed=27:27:01 avg_step=1.34s recent_step=1.78s eta=09:40:11 +2026-08-26 07:22:43 | INFO | train rank=0 | step=73960/100000 progress=74.0% loss=0.032593 lr=1.742085e-06 elapsed=27:27:13 avg_step=1.34s recent_step=1.20s eta=09:39:57 +2026-08-26 07:22:55 | INFO | train rank=0 | step=73970/100000 progress=74.0% loss=0.058624 lr=1.740831e-06 elapsed=27:27:24 avg_step=1.34s recent_step=1.14s eta=09:39:43 +2026-08-26 07:23:10 | INFO | train rank=0 | step=73980/100000 progress=74.0% loss=0.029694 lr=1.739578e-06 elapsed=27:27:40 avg_step=1.34s recent_step=1.57s eta=09:39:30 +2026-08-26 07:23:23 | INFO | train rank=0 | step=73990/100000 progress=74.0% loss=0.054062 lr=1.738324e-06 elapsed=27:27:52 avg_step=1.34s recent_step=1.23s eta=09:39:17 +2026-08-26 07:23:40 | INFO | train rank=0 | step=74000/100000 progress=74.0% loss=0.059265 lr=1.737071e-06 elapsed=27:28:10 avg_step=1.34s recent_step=1.75s eta=09:39:05 +2026-08-26 07:23:51 | INFO | train rank=0 | step=74010/100000 progress=74.0% loss=0.060791 lr=1.735819e-06 elapsed=27:28:21 avg_step=1.34s recent_step=1.13s eta=09:38:51 +2026-08-26 07:24:05 | INFO | train rank=0 | step=74020/100000 progress=74.0% loss=0.026520 lr=1.734566e-06 elapsed=27:28:35 avg_step=1.34s recent_step=1.40s eta=09:38:38 +2026-08-26 07:24:17 | INFO | train rank=0 | step=74030/100000 progress=74.0% loss=0.050507 lr=1.733314e-06 elapsed=27:28:47 avg_step=1.34s recent_step=1.18s eta=09:38:24 +2026-08-26 07:24:31 | INFO | train rank=0 | step=74040/100000 progress=74.0% loss=0.073700 lr=1.732063e-06 elapsed=27:29:01 avg_step=1.34s recent_step=1.37s eta=09:38:10 +2026-08-26 07:24:42 | INFO | train rank=0 | step=74050/100000 progress=74.1% loss=0.044693 lr=1.730811e-06 elapsed=27:29:12 avg_step=1.34s recent_step=1.17s eta=09:37:56 +2026-08-26 07:24:57 | INFO | train rank=0 | step=74060/100000 progress=74.1% loss=0.051956 lr=1.729560e-06 elapsed=27:29:27 avg_step=1.34s recent_step=1.47s eta=09:37:44 +2026-08-26 07:25:12 | INFO | train rank=0 | step=74070/100000 progress=74.1% loss=0.052399 lr=1.728310e-06 elapsed=27:29:42 avg_step=1.34s recent_step=1.47s eta=09:37:31 +2026-08-26 07:25:25 | INFO | train rank=0 | step=74080/100000 progress=74.1% loss=0.029465 lr=1.727060e-06 elapsed=27:29:55 avg_step=1.34s recent_step=1.34s eta=09:37:17 +2026-08-26 07:25:40 | INFO | train rank=0 | step=74090/100000 progress=74.1% loss=0.039200 lr=1.725810e-06 elapsed=27:30:10 avg_step=1.34s recent_step=1.49s eta=09:37:04 +2026-08-26 07:25:51 | INFO | train rank=0 | step=74100/100000 progress=74.1% loss=0.019768 lr=1.724560e-06 elapsed=27:30:21 avg_step=1.34s recent_step=1.13s eta=09:36:50 +2026-08-26 07:26:05 | INFO | train rank=0 | step=74110/100000 progress=74.1% loss=0.033573 lr=1.723311e-06 elapsed=27:30:34 avg_step=1.34s recent_step=1.31s eta=09:36:37 +2026-08-26 07:26:16 | INFO | train rank=0 | step=74120/100000 progress=74.1% loss=0.046890 lr=1.722063e-06 elapsed=27:30:46 avg_step=1.34s recent_step=1.18s eta=09:36:23 +2026-08-26 07:26:28 | INFO | train rank=0 | step=74130/100000 progress=74.1% loss=0.065002 lr=1.720814e-06 elapsed=27:30:58 avg_step=1.34s recent_step=1.19s eta=09:36:09 +2026-08-26 07:26:44 | INFO | train rank=0 | step=74140/100000 progress=74.1% loss=0.071930 lr=1.719566e-06 elapsed=27:31:13 avg_step=1.34s recent_step=1.54s eta=09:35:56 +2026-08-26 07:26:59 | INFO | train rank=0 | step=74150/100000 progress=74.2% loss=0.067993 lr=1.718319e-06 elapsed=27:31:28 avg_step=1.34s recent_step=1.50s eta=09:35:44 +2026-08-26 07:27:13 | INFO | train rank=0 | step=74160/100000 progress=74.2% loss=0.044266 lr=1.717071e-06 elapsed=27:31:43 avg_step=1.34s recent_step=1.44s eta=09:35:31 +2026-08-26 07:27:24 | INFO | train rank=0 | step=74170/100000 progress=74.2% loss=0.048325 lr=1.715824e-06 elapsed=27:31:54 avg_step=1.34s recent_step=1.16s eta=09:35:17 +2026-08-26 07:27:35 | INFO | train rank=0 | step=74180/100000 progress=74.2% loss=0.057404 lr=1.714578e-06 elapsed=27:32:05 avg_step=1.34s recent_step=1.08s eta=09:35:02 +2026-08-26 07:27:52 | INFO | train rank=0 | step=74190/100000 progress=74.2% loss=0.042465 lr=1.713331e-06 elapsed=27:32:22 avg_step=1.34s recent_step=1.64s eta=09:34:50 +2026-08-26 07:28:03 | INFO | train rank=0 | step=74200/100000 progress=74.2% loss=0.097534 lr=1.712086e-06 elapsed=27:32:33 avg_step=1.34s recent_step=1.15s eta=09:34:36 +2026-08-26 07:28:16 | INFO | train rank=0 | step=74210/100000 progress=74.2% loss=0.040924 lr=1.710840e-06 elapsed=27:32:46 avg_step=1.34s recent_step=1.31s eta=09:34:23 +2026-08-26 07:28:32 | INFO | train rank=0 | step=74220/100000 progress=74.2% loss=0.046951 lr=1.709595e-06 elapsed=27:33:02 avg_step=1.34s recent_step=1.56s eta=09:34:10 +2026-08-26 07:28:44 | INFO | train rank=0 | step=74230/100000 progress=74.2% loss=0.044296 lr=1.708350e-06 elapsed=27:33:14 avg_step=1.34s recent_step=1.23s eta=09:33:56 +2026-08-26 07:28:56 | INFO | train rank=0 | step=74240/100000 progress=74.2% loss=0.033966 lr=1.707106e-06 elapsed=27:33:25 avg_step=1.34s recent_step=1.14s eta=09:33:42 +2026-08-26 07:29:07 | INFO | train rank=0 | step=74250/100000 progress=74.2% loss=0.038116 lr=1.705862e-06 elapsed=27:33:37 avg_step=1.34s recent_step=1.18s eta=09:33:28 +2026-08-26 07:29:19 | INFO | train rank=0 | step=74260/100000 progress=74.3% loss=0.042587 lr=1.704618e-06 elapsed=27:33:49 avg_step=1.34s recent_step=1.14s eta=09:33:14 +2026-08-26 07:29:36 | INFO | train rank=0 | step=74270/100000 progress=74.3% loss=0.052933 lr=1.703375e-06 elapsed=27:34:05 avg_step=1.34s recent_step=1.68s eta=09:33:02 +2026-08-26 07:29:51 | INFO | train rank=0 | step=74280/100000 progress=74.3% loss=0.059692 lr=1.702132e-06 elapsed=27:34:21 avg_step=1.34s recent_step=1.51s eta=09:32:49 +2026-08-26 07:30:04 | INFO | train rank=0 | step=74290/100000 progress=74.3% loss=0.038269 lr=1.700889e-06 elapsed=27:34:34 avg_step=1.34s recent_step=1.36s eta=09:32:36 +2026-08-26 07:30:16 | INFO | train rank=0 | step=74300/100000 progress=74.3% loss=0.045151 lr=1.699647e-06 elapsed=27:34:46 avg_step=1.34s recent_step=1.16s eta=09:32:22 +2026-08-26 07:30:30 | INFO | train rank=0 | step=74310/100000 progress=74.3% loss=0.061127 lr=1.698405e-06 elapsed=27:35:00 avg_step=1.34s recent_step=1.40s eta=09:32:09 +2026-08-26 07:30:42 | INFO | train rank=0 | step=74320/100000 progress=74.3% loss=0.030579 lr=1.697163e-06 elapsed=27:35:12 avg_step=1.34s recent_step=1.21s eta=09:31:55 +2026-08-26 07:30:54 | INFO | train rank=0 | step=74330/100000 progress=74.3% loss=0.035934 lr=1.695922e-06 elapsed=27:35:24 avg_step=1.34s recent_step=1.20s eta=09:31:41 +2026-08-26 07:31:07 | INFO | train rank=0 | step=74340/100000 progress=74.3% loss=0.036594 lr=1.694681e-06 elapsed=27:35:37 avg_step=1.34s recent_step=1.35s eta=09:31:28 +2026-08-26 07:31:22 | INFO | train rank=0 | step=74350/100000 progress=74.4% loss=0.034454 lr=1.693441e-06 elapsed=27:35:52 avg_step=1.34s recent_step=1.47s eta=09:31:15 +2026-08-26 07:31:37 | INFO | train rank=0 | step=74360/100000 progress=74.4% loss=0.049759 lr=1.692201e-06 elapsed=27:36:07 avg_step=1.34s recent_step=1.48s eta=09:31:02 +2026-08-26 07:31:51 | INFO | train rank=0 | step=74370/100000 progress=74.4% loss=0.049583 lr=1.690961e-06 elapsed=27:36:21 avg_step=1.34s recent_step=1.46s eta=09:30:49 +2026-08-26 07:32:07 | INFO | train rank=0 | step=74380/100000 progress=74.4% loss=0.046356 lr=1.689721e-06 elapsed=27:36:37 avg_step=1.34s recent_step=1.58s eta=09:30:37 +2026-08-26 07:32:18 | INFO | train rank=0 | step=74390/100000 progress=74.4% loss=0.056107 lr=1.688482e-06 elapsed=27:36:48 avg_step=1.34s recent_step=1.11s eta=09:30:23 +2026-08-26 07:32:31 | INFO | train rank=0 | step=74400/100000 progress=74.4% loss=0.033501 lr=1.687244e-06 elapsed=27:37:01 avg_step=1.34s recent_step=1.27s eta=09:30:09 +2026-08-26 07:32:46 | INFO | train rank=0 | step=74410/100000 progress=74.4% loss=0.074524 lr=1.686005e-06 elapsed=27:37:16 avg_step=1.34s recent_step=1.46s eta=09:29:56 +2026-08-26 07:33:01 | INFO | train rank=0 | step=74420/100000 progress=74.4% loss=0.026051 lr=1.684768e-06 elapsed=27:37:31 avg_step=1.34s recent_step=1.52s eta=09:29:43 +2026-08-26 07:33:16 | INFO | train rank=0 | step=74430/100000 progress=74.4% loss=0.042145 lr=1.683530e-06 elapsed=27:37:46 avg_step=1.34s recent_step=1.50s eta=09:29:31 +2026-08-26 07:33:29 | INFO | train rank=0 | step=74440/100000 progress=74.4% loss=0.078156 lr=1.682293e-06 elapsed=27:37:59 avg_step=1.34s recent_step=1.29s eta=09:29:17 +2026-08-26 07:33:41 | INFO | train rank=0 | step=74450/100000 progress=74.5% loss=0.045807 lr=1.681056e-06 elapsed=27:38:11 avg_step=1.34s recent_step=1.23s eta=09:29:03 +2026-08-26 07:33:57 | INFO | train rank=0 | step=74460/100000 progress=74.5% loss=0.055946 lr=1.679819e-06 elapsed=27:38:27 avg_step=1.34s recent_step=1.60s eta=09:28:51 +2026-08-26 07:34:15 | INFO | train rank=0 | step=74470/100000 progress=74.5% loss=0.031685 lr=1.678583e-06 elapsed=27:38:45 avg_step=1.34s recent_step=1.81s eta=09:28:39 +2026-08-26 07:34:32 | INFO | train rank=0 | step=74480/100000 progress=74.5% loss=0.028244 lr=1.677348e-06 elapsed=27:39:02 avg_step=1.34s recent_step=1.65s eta=09:28:27 +2026-08-26 07:34:43 | INFO | train rank=0 | step=74490/100000 progress=74.5% loss=0.049088 lr=1.676112e-06 elapsed=27:39:13 avg_step=1.34s recent_step=1.13s eta=09:28:13 +2026-08-26 07:34:55 | INFO | train rank=0 | step=74500/100000 progress=74.5% loss=0.047234 lr=1.674877e-06 elapsed=27:39:25 avg_step=1.34s recent_step=1.19s eta=09:27:59 +2026-08-26 07:35:06 | INFO | train rank=0 | step=74510/100000 progress=74.5% loss=0.081238 lr=1.673642e-06 elapsed=27:39:36 avg_step=1.34s recent_step=1.14s eta=09:27:45 +2026-08-26 07:35:18 | INFO | train rank=0 | step=74520/100000 progress=74.5% loss=0.045624 lr=1.672408e-06 elapsed=27:39:48 avg_step=1.34s recent_step=1.20s eta=09:27:31 +2026-08-26 07:35:35 | INFO | train rank=0 | step=74530/100000 progress=74.5% loss=0.034042 lr=1.671174e-06 elapsed=27:40:05 avg_step=1.34s recent_step=1.64s eta=09:27:19 +2026-08-26 07:35:47 | INFO | train rank=0 | step=74540/100000 progress=74.5% loss=0.047104 lr=1.669941e-06 elapsed=27:40:17 avg_step=1.34s recent_step=1.17s eta=09:27:05 +2026-08-26 07:35:58 | INFO | train rank=0 | step=74550/100000 progress=74.6% loss=0.033615 lr=1.668707e-06 elapsed=27:40:28 avg_step=1.34s recent_step=1.18s eta=09:26:51 +2026-08-26 07:36:10 | INFO | train rank=0 | step=74560/100000 progress=74.6% loss=0.046127 lr=1.667475e-06 elapsed=27:40:40 avg_step=1.34s recent_step=1.17s eta=09:26:37 +2026-08-26 07:36:22 | INFO | train rank=0 | step=74570/100000 progress=74.6% loss=0.031616 lr=1.666242e-06 elapsed=27:40:52 avg_step=1.34s recent_step=1.17s eta=09:26:23 +2026-08-26 07:36:34 | INFO | train rank=0 | step=74580/100000 progress=74.6% loss=0.044312 lr=1.665010e-06 elapsed=27:41:04 avg_step=1.34s recent_step=1.19s eta=09:26:09 +2026-08-26 07:36:47 | INFO | train rank=0 | step=74590/100000 progress=74.6% loss=0.034515 lr=1.663778e-06 elapsed=27:41:17 avg_step=1.34s recent_step=1.36s eta=09:25:56 +2026-08-26 07:37:00 | INFO | train rank=0 | step=74600/100000 progress=74.6% loss=0.052551 lr=1.662547e-06 elapsed=27:41:30 avg_step=1.34s recent_step=1.24s eta=09:25:42 +2026-08-26 07:37:13 | INFO | train rank=0 | step=74610/100000 progress=74.6% loss=0.042694 lr=1.661316e-06 elapsed=27:41:43 avg_step=1.34s recent_step=1.33s eta=09:25:29 +2026-08-26 07:37:26 | INFO | train rank=0 | step=74620/100000 progress=74.6% loss=0.027786 lr=1.660085e-06 elapsed=27:41:56 avg_step=1.34s recent_step=1.33s eta=09:25:15 +2026-08-26 07:37:43 | INFO | train rank=0 | step=74630/100000 progress=74.6% loss=0.039978 lr=1.658855e-06 elapsed=27:42:13 avg_step=1.34s recent_step=1.66s eta=09:25:03 +2026-08-26 07:37:56 | INFO | train rank=0 | step=74640/100000 progress=74.6% loss=0.048630 lr=1.657625e-06 elapsed=27:42:26 avg_step=1.34s recent_step=1.28s eta=09:24:50 +2026-08-26 07:38:07 | INFO | train rank=0 | step=74650/100000 progress=74.7% loss=0.039352 lr=1.656395e-06 elapsed=27:42:37 avg_step=1.34s recent_step=1.17s eta=09:24:36 +2026-08-26 07:38:24 | INFO | train rank=0 | step=74660/100000 progress=74.7% loss=0.028305 lr=1.655166e-06 elapsed=27:42:54 avg_step=1.34s recent_step=1.62s eta=09:24:23 +2026-08-26 07:38:38 | INFO | train rank=0 | step=74670/100000 progress=74.7% loss=0.052429 lr=1.653937e-06 elapsed=27:43:08 avg_step=1.34s recent_step=1.45s eta=09:24:10 +2026-08-26 07:38:53 | INFO | train rank=0 | step=74680/100000 progress=74.7% loss=0.054108 lr=1.652709e-06 elapsed=27:43:23 avg_step=1.34s recent_step=1.46s eta=09:23:57 +2026-08-26 07:39:09 | INFO | train rank=0 | step=74690/100000 progress=74.7% loss=0.056503 lr=1.651481e-06 elapsed=27:43:39 avg_step=1.34s recent_step=1.61s eta=09:23:45 +2026-08-26 07:39:24 | INFO | train rank=0 | step=74700/100000 progress=74.7% loss=0.077667 lr=1.650253e-06 elapsed=27:43:54 avg_step=1.34s recent_step=1.53s eta=09:23:32 +2026-08-26 07:39:38 | INFO | train rank=0 | step=74710/100000 progress=74.7% loss=0.049377 lr=1.649026e-06 elapsed=27:44:08 avg_step=1.34s recent_step=1.36s eta=09:23:19 +2026-08-26 07:39:49 | INFO | train rank=0 | step=74720/100000 progress=74.7% loss=0.041740 lr=1.647799e-06 elapsed=27:44:19 avg_step=1.34s recent_step=1.16s eta=09:23:05 +2026-08-26 07:40:03 | INFO | train rank=0 | step=74730/100000 progress=74.7% loss=0.048248 lr=1.646572e-06 elapsed=27:44:33 avg_step=1.34s recent_step=1.36s eta=09:22:52 +2026-08-26 07:40:15 | INFO | train rank=0 | step=74740/100000 progress=74.7% loss=0.061096 lr=1.645346e-06 elapsed=27:44:45 avg_step=1.34s recent_step=1.25s eta=09:22:38 +2026-08-26 07:40:32 | INFO | train rank=0 | step=74750/100000 progress=74.8% loss=0.028519 lr=1.644120e-06 elapsed=27:45:02 avg_step=1.34s recent_step=1.63s eta=09:22:26 +2026-08-26 07:40:43 | INFO | train rank=0 | step=74760/100000 progress=74.8% loss=0.044662 lr=1.642894e-06 elapsed=27:45:13 avg_step=1.34s recent_step=1.17s eta=09:22:12 +2026-08-26 07:40:59 | INFO | train rank=0 | step=74770/100000 progress=74.8% loss=0.044052 lr=1.641669e-06 elapsed=27:45:29 avg_step=1.34s recent_step=1.57s eta=09:21:59 +2026-08-26 07:41:11 | INFO | train rank=0 | step=74780/100000 progress=74.8% loss=0.033890 lr=1.640444e-06 elapsed=27:45:41 avg_step=1.34s recent_step=1.18s eta=09:21:45 +2026-08-26 07:41:24 | INFO | train rank=0 | step=74790/100000 progress=74.8% loss=0.054077 lr=1.639220e-06 elapsed=27:45:54 avg_step=1.34s recent_step=1.29s eta=09:21:32 +2026-08-26 07:41:35 | INFO | train rank=0 | step=74800/100000 progress=74.8% loss=0.046478 lr=1.637996e-06 elapsed=27:46:05 avg_step=1.34s recent_step=1.12s eta=09:21:18 +2026-08-26 07:41:46 | INFO | train rank=0 | step=74810/100000 progress=74.8% loss=0.059875 lr=1.636772e-06 elapsed=27:46:16 avg_step=1.34s recent_step=1.12s eta=09:21:04 +2026-08-26 07:42:05 | INFO | train rank=0 | step=74820/100000 progress=74.8% loss=0.023010 lr=1.635549e-06 elapsed=27:46:35 avg_step=1.34s recent_step=1.84s eta=09:20:52 +2026-08-26 07:42:17 | INFO | train rank=0 | step=74830/100000 progress=74.8% loss=0.035843 lr=1.634326e-06 elapsed=27:46:47 avg_step=1.34s recent_step=1.20s eta=09:20:38 +2026-08-26 07:42:30 | INFO | train rank=0 | step=74840/100000 progress=74.8% loss=0.030296 lr=1.633103e-06 elapsed=27:47:00 avg_step=1.34s recent_step=1.31s eta=09:20:25 +2026-08-26 07:42:43 | INFO | train rank=0 | step=74850/100000 progress=74.9% loss=0.041656 lr=1.631881e-06 elapsed=27:47:13 avg_step=1.34s recent_step=1.34s eta=09:20:11 +2026-08-26 07:42:55 | INFO | train rank=0 | step=74860/100000 progress=74.9% loss=0.051987 lr=1.630659e-06 elapsed=27:47:25 avg_step=1.34s recent_step=1.20s eta=09:19:58 +2026-08-26 07:43:17 | INFO | train rank=0 | step=74870/100000 progress=74.9% loss=0.061600 lr=1.629438e-06 elapsed=27:47:47 avg_step=1.34s recent_step=2.20s eta=09:19:47 +2026-08-26 07:43:29 | INFO | train rank=0 | step=74880/100000 progress=74.9% loss=0.048653 lr=1.628217e-06 elapsed=27:47:59 avg_step=1.34s recent_step=1.16s eta=09:19:33 +2026-08-26 07:43:42 | INFO | train rank=0 | step=74890/100000 progress=74.9% loss=0.041584 lr=1.626996e-06 elapsed=27:48:12 avg_step=1.34s recent_step=1.29s eta=09:19:20 +2026-08-26 07:43:53 | INFO | train rank=0 | step=74900/100000 progress=74.9% loss=0.051208 lr=1.625776e-06 elapsed=27:48:23 avg_step=1.34s recent_step=1.15s eta=09:19:06 +2026-08-26 07:44:07 | INFO | train rank=0 | step=74910/100000 progress=74.9% loss=0.057556 lr=1.624556e-06 elapsed=27:48:36 avg_step=1.34s recent_step=1.34s eta=09:18:52 +2026-08-26 07:44:18 | INFO | train rank=0 | step=74920/100000 progress=74.9% loss=0.022842 lr=1.623336e-06 elapsed=27:48:48 avg_step=1.34s recent_step=1.18s eta=09:18:38 +2026-08-26 07:44:32 | INFO | train rank=0 | step=74930/100000 progress=74.9% loss=0.027039 lr=1.622117e-06 elapsed=27:49:02 avg_step=1.34s recent_step=1.41s eta=09:18:25 +2026-08-26 07:44:48 | INFO | train rank=0 | step=74940/100000 progress=74.9% loss=0.028625 lr=1.620898e-06 elapsed=27:49:17 avg_step=1.34s recent_step=1.51s eta=09:18:12 +2026-08-26 07:44:59 | INFO | train rank=0 | step=74950/100000 progress=75.0% loss=0.065170 lr=1.619679e-06 elapsed=27:49:29 avg_step=1.34s recent_step=1.16s eta=09:17:58 +2026-08-26 07:45:14 | INFO | train rank=0 | step=74960/100000 progress=75.0% loss=0.040466 lr=1.618461e-06 elapsed=27:49:44 avg_step=1.34s recent_step=1.49s eta=09:17:46 +2026-08-26 07:45:26 | INFO | train rank=0 | step=74970/100000 progress=75.0% loss=0.035858 lr=1.617243e-06 elapsed=27:49:56 avg_step=1.34s recent_step=1.19s eta=09:17:32 +2026-08-26 07:45:43 | INFO | train rank=0 | step=74980/100000 progress=75.0% loss=0.028030 lr=1.616026e-06 elapsed=27:50:13 avg_step=1.34s recent_step=1.67s eta=09:17:19 +2026-08-26 07:45:55 | INFO | train rank=0 | step=74990/100000 progress=75.0% loss=0.048660 lr=1.614809e-06 elapsed=27:50:25 avg_step=1.34s recent_step=1.27s eta=09:17:06 +2026-08-26 07:46:09 | INFO | train rank=0 | step=75000/100000 progress=75.0% loss=0.060364 lr=1.613592e-06 elapsed=27:50:39 avg_step=1.34s recent_step=1.34s eta=09:16:53 +2026-08-26 07:46:23 | INFO | train rank=0 | step=75010/100000 progress=75.0% loss=0.039719 lr=1.612376e-06 elapsed=27:50:53 avg_step=1.34s recent_step=1.41s eta=09:16:39 +2026-08-26 07:46:34 | INFO | train rank=0 | step=75020/100000 progress=75.0% loss=0.044601 lr=1.611160e-06 elapsed=27:51:04 avg_step=1.34s recent_step=1.15s eta=09:16:25 +2026-08-26 07:46:52 | INFO | train rank=0 | step=75030/100000 progress=75.0% loss=0.043060 lr=1.609944e-06 elapsed=27:51:22 avg_step=1.34s recent_step=1.79s eta=09:16:14 +2026-08-26 07:47:07 | INFO | train rank=0 | step=75040/100000 progress=75.0% loss=0.062012 lr=1.608729e-06 elapsed=27:51:36 avg_step=1.34s recent_step=1.44s eta=09:16:01 +2026-08-26 07:47:18 | INFO | train rank=0 | step=75050/100000 progress=75.0% loss=0.045082 lr=1.607514e-06 elapsed=27:51:48 avg_step=1.34s recent_step=1.17s eta=09:15:47 +2026-08-26 07:47:30 | INFO | train rank=0 | step=75060/100000 progress=75.1% loss=0.037689 lr=1.606300e-06 elapsed=27:52:00 avg_step=1.34s recent_step=1.17s eta=09:15:33 +2026-08-26 07:47:45 | INFO | train rank=0 | step=75070/100000 progress=75.1% loss=0.042557 lr=1.605086e-06 elapsed=27:52:14 avg_step=1.34s recent_step=1.47s eta=09:15:20 +2026-08-26 07:47:58 | INFO | train rank=0 | step=75080/100000 progress=75.1% loss=0.042084 lr=1.603872e-06 elapsed=27:52:28 avg_step=1.34s recent_step=1.31s eta=09:15:06 +2026-08-26 07:48:09 | INFO | train rank=0 | step=75090/100000 progress=75.1% loss=0.042297 lr=1.602659e-06 elapsed=27:52:39 avg_step=1.34s recent_step=1.10s eta=09:14:52 +2026-08-26 07:48:20 | INFO | train rank=0 | step=75100/100000 progress=75.1% loss=0.063889 lr=1.601446e-06 elapsed=27:52:50 avg_step=1.34s recent_step=1.16s eta=09:14:38 +2026-08-26 07:48:35 | INFO | train rank=0 | step=75110/100000 progress=75.1% loss=0.034286 lr=1.600233e-06 elapsed=27:53:05 avg_step=1.34s recent_step=1.47s eta=09:14:25 +2026-08-26 07:48:46 | INFO | train rank=0 | step=75120/100000 progress=75.1% loss=0.050636 lr=1.599021e-06 elapsed=27:53:16 avg_step=1.34s recent_step=1.13s eta=09:14:11 +2026-08-26 07:48:58 | INFO | train rank=0 | step=75130/100000 progress=75.1% loss=0.026474 lr=1.597809e-06 elapsed=27:53:28 avg_step=1.34s recent_step=1.17s eta=09:13:57 +2026-08-26 07:49:17 | INFO | train rank=0 | step=75140/100000 progress=75.1% loss=0.050308 lr=1.596598e-06 elapsed=27:53:47 avg_step=1.34s recent_step=1.87s eta=09:13:46 +2026-08-26 07:49:30 | INFO | train rank=0 | step=75150/100000 progress=75.1% loss=0.043640 lr=1.595386e-06 elapsed=27:54:00 avg_step=1.34s recent_step=1.33s eta=09:13:32 +2026-08-26 07:49:42 | INFO | train rank=0 | step=75160/100000 progress=75.2% loss=0.032150 lr=1.594176e-06 elapsed=27:54:12 avg_step=1.34s recent_step=1.17s eta=09:13:18 +2026-08-26 07:49:53 | INFO | train rank=0 | step=75170/100000 progress=75.2% loss=0.052689 lr=1.592965e-06 elapsed=27:54:22 avg_step=1.34s recent_step=1.09s eta=09:13:04 +2026-08-26 07:50:12 | INFO | train rank=0 | step=75180/100000 progress=75.2% loss=0.048599 lr=1.591755e-06 elapsed=27:54:42 avg_step=1.34s recent_step=1.94s eta=09:12:53 +2026-08-26 07:50:29 | INFO | train rank=0 | step=75190/100000 progress=75.2% loss=0.050995 lr=1.590546e-06 elapsed=27:54:59 avg_step=1.34s recent_step=1.66s eta=09:12:41 +2026-08-26 07:50:40 | INFO | train rank=0 | step=75200/100000 progress=75.2% loss=0.035706 lr=1.589336e-06 elapsed=27:55:10 avg_step=1.34s recent_step=1.16s eta=09:12:27 +2026-08-26 07:50:54 | INFO | train rank=0 | step=75210/100000 progress=75.2% loss=0.040970 lr=1.588128e-06 elapsed=27:55:24 avg_step=1.34s recent_step=1.37s eta=09:12:13 +2026-08-26 07:51:09 | INFO | train rank=0 | step=75220/100000 progress=75.2% loss=0.035172 lr=1.586919e-06 elapsed=27:55:38 avg_step=1.34s recent_step=1.46s eta=09:12:00 +2026-08-26 07:51:20 | INFO | train rank=0 | step=75230/100000 progress=75.2% loss=0.029335 lr=1.585711e-06 elapsed=27:55:50 avg_step=1.34s recent_step=1.16s eta=09:11:46 +2026-08-26 07:51:34 | INFO | train rank=0 | step=75240/100000 progress=75.2% loss=0.035919 lr=1.584503e-06 elapsed=27:56:04 avg_step=1.34s recent_step=1.36s eta=09:11:33 +2026-08-26 07:51:45 | INFO | train rank=0 | step=75250/100000 progress=75.2% loss=0.027794 lr=1.583296e-06 elapsed=27:56:15 avg_step=1.34s recent_step=1.13s eta=09:11:19 +2026-08-26 07:51:59 | INFO | train rank=0 | step=75260/100000 progress=75.3% loss=0.065384 lr=1.582089e-06 elapsed=27:56:29 avg_step=1.34s recent_step=1.38s eta=09:11:06 +2026-08-26 07:52:11 | INFO | train rank=0 | step=75270/100000 progress=75.3% loss=0.038773 lr=1.580882e-06 elapsed=27:56:41 avg_step=1.34s recent_step=1.20s eta=09:10:52 +2026-08-26 07:52:22 | INFO | train rank=0 | step=75280/100000 progress=75.3% loss=0.033295 lr=1.579676e-06 elapsed=27:56:52 avg_step=1.34s recent_step=1.13s eta=09:10:38 +2026-08-26 07:52:34 | INFO | train rank=0 | step=75290/100000 progress=75.3% loss=0.031128 lr=1.578470e-06 elapsed=27:57:03 avg_step=1.34s recent_step=1.13s eta=09:10:24 +2026-08-26 07:52:47 | INFO | train rank=0 | step=75300/100000 progress=75.3% loss=0.074066 lr=1.577264e-06 elapsed=27:57:17 avg_step=1.34s recent_step=1.38s eta=09:10:11 +2026-08-26 07:53:00 | INFO | train rank=0 | step=75310/100000 progress=75.3% loss=0.026962 lr=1.576059e-06 elapsed=27:57:30 avg_step=1.34s recent_step=1.26s eta=09:09:57 +2026-08-26 07:53:12 | INFO | train rank=0 | step=75320/100000 progress=75.3% loss=0.047119 lr=1.574855e-06 elapsed=27:57:42 avg_step=1.34s recent_step=1.18s eta=09:09:43 +2026-08-26 07:53:23 | INFO | train rank=0 | step=75330/100000 progress=75.3% loss=0.042603 lr=1.573650e-06 elapsed=27:57:53 avg_step=1.34s recent_step=1.12s eta=09:09:29 +2026-08-26 07:53:35 | INFO | train rank=0 | step=75340/100000 progress=75.3% loss=0.039902 lr=1.572446e-06 elapsed=27:58:05 avg_step=1.34s recent_step=1.19s eta=09:09:15 +2026-08-26 07:53:46 | INFO | train rank=0 | step=75350/100000 progress=75.3% loss=0.046570 lr=1.571243e-06 elapsed=27:58:16 avg_step=1.34s recent_step=1.15s eta=09:09:01 +2026-08-26 07:53:59 | INFO | train rank=0 | step=75360/100000 progress=75.4% loss=0.047783 lr=1.570039e-06 elapsed=27:58:29 avg_step=1.34s recent_step=1.31s eta=09:08:48 +2026-08-26 07:54:11 | INFO | train rank=0 | step=75370/100000 progress=75.4% loss=0.044258 lr=1.568836e-06 elapsed=27:58:41 avg_step=1.34s recent_step=1.20s eta=09:08:34 +2026-08-26 07:54:23 | INFO | train rank=0 | step=75380/100000 progress=75.4% loss=0.040943 lr=1.567634e-06 elapsed=27:58:53 avg_step=1.34s recent_step=1.16s eta=09:08:20 +2026-08-26 07:54:37 | INFO | train rank=0 | step=75390/100000 progress=75.4% loss=0.038986 lr=1.566432e-06 elapsed=27:59:07 avg_step=1.34s recent_step=1.42s eta=09:08:07 +2026-08-26 07:54:48 | INFO | train rank=0 | step=75400/100000 progress=75.4% loss=0.031677 lr=1.565230e-06 elapsed=27:59:18 avg_step=1.34s recent_step=1.12s eta=09:07:53 +2026-08-26 07:55:04 | INFO | train rank=0 | step=75410/100000 progress=75.4% loss=0.047928 lr=1.564029e-06 elapsed=27:59:34 avg_step=1.34s recent_step=1.56s eta=09:07:40 +2026-08-26 07:55:18 | INFO | train rank=0 | step=75420/100000 progress=75.4% loss=0.046249 lr=1.562828e-06 elapsed=27:59:48 avg_step=1.34s recent_step=1.42s eta=09:07:27 +2026-08-26 07:55:31 | INFO | train rank=0 | step=75430/100000 progress=75.4% loss=0.054962 lr=1.561627e-06 elapsed=28:00:01 avg_step=1.34s recent_step=1.29s eta=09:07:14 +2026-08-26 07:55:46 | INFO | train rank=0 | step=75440/100000 progress=75.4% loss=0.049088 lr=1.560427e-06 elapsed=28:00:15 avg_step=1.34s recent_step=1.46s eta=09:07:01 +2026-08-26 07:55:58 | INFO | train rank=0 | step=75450/100000 progress=75.4% loss=0.049133 lr=1.559227e-06 elapsed=28:00:28 avg_step=1.34s recent_step=1.28s eta=09:06:47 +2026-08-26 07:56:12 | INFO | train rank=0 | step=75460/100000 progress=75.5% loss=0.037964 lr=1.558027e-06 elapsed=28:00:42 avg_step=1.34s recent_step=1.36s eta=09:06:34 +2026-08-26 07:56:32 | INFO | train rank=0 | step=75470/100000 progress=75.5% loss=0.035301 lr=1.556828e-06 elapsed=28:01:02 avg_step=1.34s recent_step=2.01s eta=09:06:23 +2026-08-26 07:56:47 | INFO | train rank=0 | step=75480/100000 progress=75.5% loss=0.039719 lr=1.555629e-06 elapsed=28:01:17 avg_step=1.34s recent_step=1.49s eta=09:06:10 +2026-08-26 07:57:01 | INFO | train rank=0 | step=75490/100000 progress=75.5% loss=0.034660 lr=1.554431e-06 elapsed=28:01:31 avg_step=1.34s recent_step=1.45s eta=09:05:57 +2026-08-26 07:57:13 | INFO | train rank=0 | step=75500/100000 progress=75.5% loss=0.037796 lr=1.553233e-06 elapsed=28:01:43 avg_step=1.34s recent_step=1.15s eta=09:05:43 +2026-08-26 07:57:28 | INFO | train rank=0 | step=75510/100000 progress=75.5% loss=0.116455 lr=1.552035e-06 elapsed=28:01:58 avg_step=1.34s recent_step=1.50s eta=09:05:30 +2026-08-26 07:57:40 | INFO | train rank=0 | step=75520/100000 progress=75.5% loss=0.038376 lr=1.550838e-06 elapsed=28:02:10 avg_step=1.34s recent_step=1.21s eta=09:05:16 +2026-08-26 07:57:54 | INFO | train rank=0 | step=75530/100000 progress=75.5% loss=0.037003 lr=1.549641e-06 elapsed=28:02:24 avg_step=1.34s recent_step=1.40s eta=09:05:03 +2026-08-26 07:58:06 | INFO | train rank=0 | step=75540/100000 progress=75.5% loss=0.073517 lr=1.548445e-06 elapsed=28:02:36 avg_step=1.34s recent_step=1.20s eta=09:04:49 +2026-08-26 07:58:19 | INFO | train rank=0 | step=75550/100000 progress=75.5% loss=0.045944 lr=1.547249e-06 elapsed=28:02:49 avg_step=1.34s recent_step=1.31s eta=09:04:36 +2026-08-26 07:58:35 | INFO | train rank=0 | step=75560/100000 progress=75.6% loss=0.037567 lr=1.546053e-06 elapsed=28:03:05 avg_step=1.34s recent_step=1.58s eta=09:04:23 +2026-08-26 07:58:51 | INFO | train rank=0 | step=75570/100000 progress=75.6% loss=0.041046 lr=1.544857e-06 elapsed=28:03:21 avg_step=1.34s recent_step=1.59s eta=09:04:11 +2026-08-26 07:59:06 | INFO | train rank=0 | step=75580/100000 progress=75.6% loss=0.051315 lr=1.543663e-06 elapsed=28:03:36 avg_step=1.34s recent_step=1.56s eta=09:03:58 +2026-08-26 07:59:19 | INFO | train rank=0 | step=75590/100000 progress=75.6% loss=0.054321 lr=1.542468e-06 elapsed=28:03:49 avg_step=1.34s recent_step=1.28s eta=09:03:45 +2026-08-26 07:59:32 | INFO | train rank=0 | step=75600/100000 progress=75.6% loss=0.036148 lr=1.541274e-06 elapsed=28:04:02 avg_step=1.34s recent_step=1.25s eta=09:03:31 +2026-08-26 07:59:43 | INFO | train rank=0 | step=75610/100000 progress=75.6% loss=0.044037 lr=1.540080e-06 elapsed=28:04:13 avg_step=1.34s recent_step=1.13s eta=09:03:17 +2026-08-26 08:00:00 | INFO | train rank=0 | step=75620/100000 progress=75.6% loss=0.035004 lr=1.538886e-06 elapsed=28:04:30 avg_step=1.34s recent_step=1.72s eta=09:03:05 +2026-08-26 08:00:13 | INFO | train rank=0 | step=75630/100000 progress=75.6% loss=0.048233 lr=1.537693e-06 elapsed=28:04:43 avg_step=1.34s recent_step=1.31s eta=09:02:51 +2026-08-26 08:00:31 | INFO | train rank=0 | step=75640/100000 progress=75.6% loss=0.045395 lr=1.536501e-06 elapsed=28:05:01 avg_step=1.34s recent_step=1.76s eta=09:02:39 +2026-08-26 08:00:43 | INFO | train rank=0 | step=75650/100000 progress=75.6% loss=0.025909 lr=1.535308e-06 elapsed=28:05:12 avg_step=1.34s recent_step=1.16s eta=09:02:25 +2026-08-26 08:00:58 | INFO | train rank=0 | step=75660/100000 progress=75.7% loss=0.044617 lr=1.534116e-06 elapsed=28:05:28 avg_step=1.34s recent_step=1.51s eta=09:02:13 +2026-08-26 08:01:09 | INFO | train rank=0 | step=75670/100000 progress=75.7% loss=0.037140 lr=1.532925e-06 elapsed=28:05:39 avg_step=1.34s recent_step=1.15s eta=09:01:59 +2026-08-26 08:01:20 | INFO | train rank=0 | step=75680/100000 progress=75.7% loss=0.036163 lr=1.531733e-06 elapsed=28:05:50 avg_step=1.34s recent_step=1.13s eta=09:01:45 +2026-08-26 08:01:35 | INFO | train rank=0 | step=75690/100000 progress=75.7% loss=0.031898 lr=1.530543e-06 elapsed=28:06:05 avg_step=1.34s recent_step=1.46s eta=09:01:32 +2026-08-26 08:01:50 | INFO | train rank=0 | step=75700/100000 progress=75.7% loss=0.057205 lr=1.529352e-06 elapsed=28:06:20 avg_step=1.34s recent_step=1.51s eta=09:01:19 +2026-08-26 08:02:03 | INFO | train rank=0 | step=75710/100000 progress=75.7% loss=0.033493 lr=1.528162e-06 elapsed=28:06:33 avg_step=1.34s recent_step=1.28s eta=09:01:05 +2026-08-26 08:02:16 | INFO | train rank=0 | step=75720/100000 progress=75.7% loss=0.060089 lr=1.526972e-06 elapsed=28:06:46 avg_step=1.34s recent_step=1.36s eta=09:00:52 +2026-08-26 08:02:32 | INFO | train rank=0 | step=75730/100000 progress=75.7% loss=0.048172 lr=1.525783e-06 elapsed=28:07:02 avg_step=1.34s recent_step=1.60s eta=09:00:39 +2026-08-26 08:02:44 | INFO | train rank=0 | step=75740/100000 progress=75.7% loss=0.044418 lr=1.524594e-06 elapsed=28:07:14 avg_step=1.34s recent_step=1.13s eta=09:00:25 +2026-08-26 08:02:58 | INFO | train rank=0 | step=75750/100000 progress=75.8% loss=0.042442 lr=1.523406e-06 elapsed=28:07:28 avg_step=1.34s recent_step=1.43s eta=09:00:12 +2026-08-26 08:03:12 | INFO | train rank=0 | step=75760/100000 progress=75.8% loss=0.030952 lr=1.522218e-06 elapsed=28:07:42 avg_step=1.34s recent_step=1.41s eta=08:59:59 +2026-08-26 08:03:23 | INFO | train rank=0 | step=75770/100000 progress=75.8% loss=0.030594 lr=1.521030e-06 elapsed=28:07:53 avg_step=1.34s recent_step=1.12s eta=08:59:45 +2026-08-26 08:03:37 | INFO | train rank=0 | step=75780/100000 progress=75.8% loss=0.064331 lr=1.519842e-06 elapsed=28:08:07 avg_step=1.34s recent_step=1.35s eta=08:59:32 +2026-08-26 08:03:51 | INFO | train rank=0 | step=75790/100000 progress=75.8% loss=0.054840 lr=1.518655e-06 elapsed=28:08:20 avg_step=1.34s recent_step=1.37s eta=08:59:19 +2026-08-26 08:04:04 | INFO | train rank=0 | step=75800/100000 progress=75.8% loss=0.055817 lr=1.517469e-06 elapsed=28:08:34 avg_step=1.34s recent_step=1.39s eta=08:59:05 +2026-08-26 08:04:16 | INFO | train rank=0 | step=75810/100000 progress=75.8% loss=0.063980 lr=1.516282e-06 elapsed=28:08:46 avg_step=1.34s recent_step=1.14s eta=08:58:51 +2026-08-26 08:04:35 | INFO | train rank=0 | step=75820/100000 progress=75.8% loss=0.052658 lr=1.515097e-06 elapsed=28:09:05 avg_step=1.34s recent_step=1.96s eta=08:58:40 +2026-08-26 08:04:47 | INFO | train rank=0 | step=75830/100000 progress=75.8% loss=0.045807 lr=1.513911e-06 elapsed=28:09:17 avg_step=1.34s recent_step=1.15s eta=08:58:26 +2026-08-26 08:04:59 | INFO | train rank=0 | step=75840/100000 progress=75.8% loss=0.051682 lr=1.512726e-06 elapsed=28:09:29 avg_step=1.34s recent_step=1.19s eta=08:58:12 +2026-08-26 08:05:14 | INFO | train rank=0 | step=75850/100000 progress=75.8% loss=0.040451 lr=1.511541e-06 elapsed=28:09:44 avg_step=1.34s recent_step=1.54s eta=08:58:00 +2026-08-26 08:05:27 | INFO | train rank=0 | step=75860/100000 progress=75.9% loss=0.045898 lr=1.510357e-06 elapsed=28:09:57 avg_step=1.34s recent_step=1.31s eta=08:57:46 +2026-08-26 08:05:39 | INFO | train rank=0 | step=75870/100000 progress=75.9% loss=0.046539 lr=1.509173e-06 elapsed=28:10:09 avg_step=1.34s recent_step=1.20s eta=08:57:32 +2026-08-26 08:05:51 | INFO | train rank=0 | step=75880/100000 progress=75.9% loss=0.047455 lr=1.507989e-06 elapsed=28:10:21 avg_step=1.34s recent_step=1.14s eta=08:57:18 +2026-08-26 08:06:03 | INFO | train rank=0 | step=75890/100000 progress=75.9% loss=0.050873 lr=1.506806e-06 elapsed=28:10:32 avg_step=1.34s recent_step=1.18s eta=08:57:04 +2026-08-26 08:06:15 | INFO | train rank=0 | step=75900/100000 progress=75.9% loss=0.035110 lr=1.505623e-06 elapsed=28:10:45 avg_step=1.34s recent_step=1.26s eta=08:56:51 +2026-08-26 08:06:28 | INFO | train rank=0 | step=75910/100000 progress=75.9% loss=0.042877 lr=1.504441e-06 elapsed=28:10:58 avg_step=1.34s recent_step=1.26s eta=08:56:37 +2026-08-26 08:06:43 | INFO | train rank=0 | step=75920/100000 progress=75.9% loss=0.031799 lr=1.503259e-06 elapsed=28:11:13 avg_step=1.34s recent_step=1.51s eta=08:56:24 +2026-08-26 08:06:57 | INFO | train rank=0 | step=75930/100000 progress=75.9% loss=0.067566 lr=1.502077e-06 elapsed=28:11:27 avg_step=1.34s recent_step=1.43s eta=08:56:11 +2026-08-26 08:07:12 | INFO | train rank=0 | step=75940/100000 progress=75.9% loss=0.043823 lr=1.500896e-06 elapsed=28:11:41 avg_step=1.34s recent_step=1.43s eta=08:55:58 +2026-08-26 08:07:31 | INFO | train rank=0 | step=75950/100000 progress=75.9% loss=0.043198 lr=1.499715e-06 elapsed=28:12:01 avg_step=1.34s recent_step=1.93s eta=08:55:47 +2026-08-26 08:07:42 | INFO | train rank=0 | step=75960/100000 progress=76.0% loss=0.066788 lr=1.498534e-06 elapsed=28:12:12 avg_step=1.34s recent_step=1.13s eta=08:55:33 +2026-08-26 08:07:54 | INFO | train rank=0 | step=75970/100000 progress=76.0% loss=0.037598 lr=1.497354e-06 elapsed=28:12:24 avg_step=1.34s recent_step=1.19s eta=08:55:19 +2026-08-26 08:08:05 | INFO | train rank=0 | step=75980/100000 progress=76.0% loss=0.060852 lr=1.496174e-06 elapsed=28:12:35 avg_step=1.34s recent_step=1.13s eta=08:55:05 +2026-08-26 08:08:17 | INFO | train rank=0 | step=75990/100000 progress=76.0% loss=0.052704 lr=1.494995e-06 elapsed=28:12:47 avg_step=1.34s recent_step=1.13s eta=08:54:51 +2026-08-26 08:08:29 | INFO | train rank=0 | step=76000/100000 progress=76.0% loss=0.034592 lr=1.493816e-06 elapsed=28:12:59 avg_step=1.34s recent_step=1.22s eta=08:54:37 +2026-08-26 08:08:47 | INFO | train rank=0 | step=76010/100000 progress=76.0% loss=0.049255 lr=1.492637e-06 elapsed=28:13:17 avg_step=1.34s recent_step=1.83s eta=08:54:25 +2026-08-26 08:08:59 | INFO | train rank=0 | step=76020/100000 progress=76.0% loss=0.050903 lr=1.491459e-06 elapsed=28:13:29 avg_step=1.34s recent_step=1.17s eta=08:54:11 +2026-08-26 08:09:13 | INFO | train rank=0 | step=76030/100000 progress=76.0% loss=0.043129 lr=1.490281e-06 elapsed=28:13:43 avg_step=1.34s recent_step=1.42s eta=08:53:58 +2026-08-26 08:09:31 | INFO | train rank=0 | step=76040/100000 progress=76.0% loss=0.027023 lr=1.489104e-06 elapsed=28:14:01 avg_step=1.34s recent_step=1.76s eta=08:53:46 +2026-08-26 08:09:43 | INFO | train rank=0 | step=76050/100000 progress=76.0% loss=0.023033 lr=1.487927e-06 elapsed=28:14:13 avg_step=1.34s recent_step=1.24s eta=08:53:33 +2026-08-26 08:09:57 | INFO | train rank=0 | step=76060/100000 progress=76.1% loss=0.051117 lr=1.486750e-06 elapsed=28:14:27 avg_step=1.34s recent_step=1.40s eta=08:53:20 +2026-08-26 08:10:09 | INFO | train rank=0 | step=76070/100000 progress=76.1% loss=0.043503 lr=1.485574e-06 elapsed=28:14:38 avg_step=1.34s recent_step=1.14s eta=08:53:06 +2026-08-26 08:10:20 | INFO | train rank=0 | step=76080/100000 progress=76.1% loss=0.032845 lr=1.484398e-06 elapsed=28:14:50 avg_step=1.34s recent_step=1.19s eta=08:52:52 +2026-08-26 08:10:32 | INFO | train rank=0 | step=76090/100000 progress=76.1% loss=0.065643 lr=1.483222e-06 elapsed=28:15:02 avg_step=1.34s recent_step=1.14s eta=08:52:38 +2026-08-26 08:10:45 | INFO | train rank=0 | step=76100/100000 progress=76.1% loss=0.044998 lr=1.482047e-06 elapsed=28:15:15 avg_step=1.34s recent_step=1.29s eta=08:52:24 +2026-08-26 08:11:00 | INFO | train rank=0 | step=76110/100000 progress=76.1% loss=0.054504 lr=1.480872e-06 elapsed=28:15:30 avg_step=1.34s recent_step=1.52s eta=08:52:11 +2026-08-26 08:11:13 | INFO | train rank=0 | step=76120/100000 progress=76.1% loss=0.050140 lr=1.479698e-06 elapsed=28:15:43 avg_step=1.34s recent_step=1.35s eta=08:51:58 +2026-08-26 08:11:25 | INFO | train rank=0 | step=76130/100000 progress=76.1% loss=0.047806 lr=1.478524e-06 elapsed=28:15:55 avg_step=1.34s recent_step=1.17s eta=08:51:44 +2026-08-26 08:11:37 | INFO | train rank=0 | step=76140/100000 progress=76.1% loss=0.042839 lr=1.477350e-06 elapsed=28:16:07 avg_step=1.34s recent_step=1.17s eta=08:51:30 +2026-08-26 08:11:51 | INFO | train rank=0 | step=76150/100000 progress=76.1% loss=0.084167 lr=1.476177e-06 elapsed=28:16:20 avg_step=1.34s recent_step=1.37s eta=08:51:17 +2026-08-26 08:12:07 | INFO | train rank=0 | step=76160/100000 progress=76.2% loss=0.055115 lr=1.475004e-06 elapsed=28:16:37 avg_step=1.34s recent_step=1.62s eta=08:51:05 +2026-08-26 08:12:19 | INFO | train rank=0 | step=76170/100000 progress=76.2% loss=0.038361 lr=1.473832e-06 elapsed=28:16:49 avg_step=1.34s recent_step=1.25s eta=08:50:51 +2026-08-26 08:12:33 | INFO | train rank=0 | step=76180/100000 progress=76.2% loss=0.039841 lr=1.472660e-06 elapsed=28:17:03 avg_step=1.34s recent_step=1.34s eta=08:50:38 +2026-08-26 08:12:49 | INFO | train rank=0 | step=76190/100000 progress=76.2% loss=0.044373 lr=1.471488e-06 elapsed=28:17:19 avg_step=1.34s recent_step=1.66s eta=08:50:25 +2026-08-26 08:13:02 | INFO | train rank=0 | step=76200/100000 progress=76.2% loss=0.049362 lr=1.470317e-06 elapsed=28:17:32 avg_step=1.34s recent_step=1.25s eta=08:50:12 +2026-08-26 08:13:15 | INFO | train rank=0 | step=76210/100000 progress=76.2% loss=0.036263 lr=1.469146e-06 elapsed=28:17:45 avg_step=1.34s recent_step=1.30s eta=08:49:58 +2026-08-26 08:13:28 | INFO | train rank=0 | step=76220/100000 progress=76.2% loss=0.050323 lr=1.467975e-06 elapsed=28:17:58 avg_step=1.34s recent_step=1.34s eta=08:49:45 +2026-08-26 08:13:44 | INFO | train rank=0 | step=76230/100000 progress=76.2% loss=0.041946 lr=1.466805e-06 elapsed=28:18:14 avg_step=1.34s recent_step=1.59s eta=08:49:32 +2026-08-26 08:13:57 | INFO | train rank=0 | step=76240/100000 progress=76.2% loss=0.045166 lr=1.465635e-06 elapsed=28:18:27 avg_step=1.34s recent_step=1.30s eta=08:49:19 +2026-08-26 08:14:12 | INFO | train rank=0 | step=76250/100000 progress=76.2% loss=0.065842 lr=1.464466e-06 elapsed=28:18:42 avg_step=1.34s recent_step=1.52s eta=08:49:06 +2026-08-26 08:14:24 | INFO | train rank=0 | step=76260/100000 progress=76.3% loss=0.077576 lr=1.463297e-06 elapsed=28:18:54 avg_step=1.34s recent_step=1.19s eta=08:48:52 +2026-08-26 08:14:36 | INFO | train rank=0 | step=76270/100000 progress=76.3% loss=0.035934 lr=1.462129e-06 elapsed=28:19:06 avg_step=1.34s recent_step=1.18s eta=08:48:38 +2026-08-26 08:14:51 | INFO | train rank=0 | step=76280/100000 progress=76.3% loss=0.047729 lr=1.460960e-06 elapsed=28:19:21 avg_step=1.34s recent_step=1.55s eta=08:48:25 +2026-08-26 08:15:03 | INFO | train rank=0 | step=76290/100000 progress=76.3% loss=0.035202 lr=1.459792e-06 elapsed=28:19:33 avg_step=1.34s recent_step=1.19s eta=08:48:12 +2026-08-26 08:15:15 | INFO | train rank=0 | step=76300/100000 progress=76.3% loss=0.098816 lr=1.458625e-06 elapsed=28:19:45 avg_step=1.34s recent_step=1.18s eta=08:47:58 +2026-08-26 08:15:31 | INFO | train rank=0 | step=76310/100000 progress=76.3% loss=0.044838 lr=1.457458e-06 elapsed=28:20:01 avg_step=1.34s recent_step=1.57s eta=08:47:45 +2026-08-26 08:15:42 | INFO | train rank=0 | step=76320/100000 progress=76.3% loss=0.037552 lr=1.456291e-06 elapsed=28:20:12 avg_step=1.34s recent_step=1.17s eta=08:47:31 +2026-08-26 08:15:54 | INFO | train rank=0 | step=76330/100000 progress=76.3% loss=0.071259 lr=1.455125e-06 elapsed=28:20:24 avg_step=1.34s recent_step=1.15s eta=08:47:17 +2026-08-26 08:16:08 | INFO | train rank=0 | step=76340/100000 progress=76.3% loss=0.039856 lr=1.453959e-06 elapsed=28:20:38 avg_step=1.34s recent_step=1.38s eta=08:47:04 +2026-08-26 08:16:23 | INFO | train rank=0 | step=76350/100000 progress=76.3% loss=0.058716 lr=1.452794e-06 elapsed=28:20:53 avg_step=1.34s recent_step=1.55s eta=08:46:51 +2026-08-26 08:16:35 | INFO | train rank=0 | step=76360/100000 progress=76.4% loss=0.050629 lr=1.451629e-06 elapsed=28:21:05 avg_step=1.34s recent_step=1.20s eta=08:46:38 +2026-08-26 08:16:50 | INFO | train rank=0 | step=76370/100000 progress=76.4% loss=0.053802 lr=1.450464e-06 elapsed=28:21:20 avg_step=1.34s recent_step=1.46s eta=08:46:25 +2026-08-26 08:17:03 | INFO | train rank=0 | step=76380/100000 progress=76.4% loss=0.061615 lr=1.449299e-06 elapsed=28:21:33 avg_step=1.34s recent_step=1.29s eta=08:46:11 +2026-08-26 08:17:16 | INFO | train rank=0 | step=76390/100000 progress=76.4% loss=0.038727 lr=1.448136e-06 elapsed=28:21:46 avg_step=1.34s recent_step=1.35s eta=08:45:58 +2026-08-26 08:17:31 | INFO | train rank=0 | step=76400/100000 progress=76.4% loss=0.063690 lr=1.446972e-06 elapsed=28:22:01 avg_step=1.34s recent_step=1.44s eta=08:45:45 +2026-08-26 08:17:46 | INFO | train rank=0 | step=76410/100000 progress=76.4% loss=0.028198 lr=1.445809e-06 elapsed=28:22:16 avg_step=1.34s recent_step=1.56s eta=08:45:32 +2026-08-26 08:18:00 | INFO | train rank=0 | step=76420/100000 progress=76.4% loss=0.072113 lr=1.444646e-06 elapsed=28:22:30 avg_step=1.34s recent_step=1.35s eta=08:45:19 +2026-08-26 08:18:14 | INFO | train rank=0 | step=76430/100000 progress=76.4% loss=0.038879 lr=1.443484e-06 elapsed=28:22:44 avg_step=1.34s recent_step=1.43s eta=08:45:06 +2026-08-26 08:18:25 | INFO | train rank=0 | step=76440/100000 progress=76.4% loss=0.043823 lr=1.442322e-06 elapsed=28:22:55 avg_step=1.34s recent_step=1.14s eta=08:44:52 +2026-08-26 08:18:40 | INFO | train rank=0 | step=76450/100000 progress=76.4% loss=0.064728 lr=1.441160e-06 elapsed=28:23:10 avg_step=1.34s recent_step=1.43s eta=08:44:39 +2026-08-26 08:18:54 | INFO | train rank=0 | step=76460/100000 progress=76.5% loss=0.044533 lr=1.439999e-06 elapsed=28:23:24 avg_step=1.34s recent_step=1.41s eta=08:44:25 +2026-08-26 08:19:07 | INFO | train rank=0 | step=76470/100000 progress=76.5% loss=0.028484 lr=1.438838e-06 elapsed=28:23:37 avg_step=1.34s recent_step=1.36s eta=08:44:12 +2026-08-26 08:19:21 | INFO | train rank=0 | step=76480/100000 progress=76.5% loss=0.045944 lr=1.437677e-06 elapsed=28:23:51 avg_step=1.34s recent_step=1.39s eta=08:43:59 +2026-08-26 08:19:33 | INFO | train rank=0 | step=76490/100000 progress=76.5% loss=0.054855 lr=1.436517e-06 elapsed=28:24:03 avg_step=1.34s recent_step=1.18s eta=08:43:45 +2026-08-26 08:19:44 | INFO | train rank=0 | step=76500/100000 progress=76.5% loss=0.045868 lr=1.435358e-06 elapsed=28:24:14 avg_step=1.34s recent_step=1.13s eta=08:43:31 +2026-08-26 08:20:01 | INFO | train rank=0 | step=76510/100000 progress=76.5% loss=0.051254 lr=1.434198e-06 elapsed=28:24:31 avg_step=1.34s recent_step=1.70s eta=08:43:19 +2026-08-26 08:20:16 | INFO | train rank=0 | step=76520/100000 progress=76.5% loss=0.066437 lr=1.433040e-06 elapsed=28:24:46 avg_step=1.34s recent_step=1.50s eta=08:43:06 +2026-08-26 08:20:31 | INFO | train rank=0 | step=76530/100000 progress=76.5% loss=0.039452 lr=1.431881e-06 elapsed=28:25:01 avg_step=1.34s recent_step=1.44s eta=08:42:53 +2026-08-26 08:20:43 | INFO | train rank=0 | step=76540/100000 progress=76.5% loss=0.040741 lr=1.430723e-06 elapsed=28:25:13 avg_step=1.34s recent_step=1.19s eta=08:42:39 +2026-08-26 08:20:56 | INFO | train rank=0 | step=76550/100000 progress=76.5% loss=0.034195 lr=1.429565e-06 elapsed=28:25:26 avg_step=1.34s recent_step=1.32s eta=08:42:26 +2026-08-26 08:21:10 | INFO | train rank=0 | step=76560/100000 progress=76.6% loss=0.034882 lr=1.428408e-06 elapsed=28:25:40 avg_step=1.34s recent_step=1.40s eta=08:42:13 +2026-08-26 08:21:21 | INFO | train rank=0 | step=76570/100000 progress=76.6% loss=0.028831 lr=1.427251e-06 elapsed=28:25:51 avg_step=1.34s recent_step=1.12s eta=08:41:59 +2026-08-26 08:21:33 | INFO | train rank=0 | step=76580/100000 progress=76.6% loss=0.057297 lr=1.426094e-06 elapsed=28:26:03 avg_step=1.34s recent_step=1.20s eta=08:41:45 +2026-08-26 08:21:49 | INFO | train rank=0 | step=76590/100000 progress=76.6% loss=0.066956 lr=1.424938e-06 elapsed=28:26:19 avg_step=1.34s recent_step=1.61s eta=08:41:32 +2026-08-26 08:22:05 | INFO | train rank=0 | step=76600/100000 progress=76.6% loss=0.039558 lr=1.423783e-06 elapsed=28:26:35 avg_step=1.34s recent_step=1.57s eta=08:41:20 +2026-08-26 08:22:19 | INFO | train rank=0 | step=76610/100000 progress=76.6% loss=0.027672 lr=1.422627e-06 elapsed=28:26:49 avg_step=1.34s recent_step=1.37s eta=08:41:06 +2026-08-26 08:22:30 | INFO | train rank=0 | step=76620/100000 progress=76.6% loss=0.084167 lr=1.421472e-06 elapsed=28:27:00 avg_step=1.34s recent_step=1.16s eta=08:40:52 +2026-08-26 08:22:42 | INFO | train rank=0 | step=76630/100000 progress=76.6% loss=0.038788 lr=1.420318e-06 elapsed=28:27:12 avg_step=1.34s recent_step=1.16s eta=08:40:38 +2026-08-26 08:22:55 | INFO | train rank=0 | step=76640/100000 progress=76.6% loss=0.045013 lr=1.419163e-06 elapsed=28:27:25 avg_step=1.34s recent_step=1.35s eta=08:40:25 +2026-08-26 08:23:07 | INFO | train rank=0 | step=76650/100000 progress=76.6% loss=0.041885 lr=1.418010e-06 elapsed=28:27:36 avg_step=1.34s recent_step=1.11s eta=08:40:11 +2026-08-26 08:23:18 | INFO | train rank=0 | step=76660/100000 progress=76.7% loss=0.035454 lr=1.416856e-06 elapsed=28:27:48 avg_step=1.34s recent_step=1.11s eta=08:39:57 +2026-08-26 08:23:29 | INFO | train rank=0 | step=76670/100000 progress=76.7% loss=0.069855 lr=1.415703e-06 elapsed=28:27:59 avg_step=1.34s recent_step=1.18s eta=08:39:43 +2026-08-26 08:23:44 | INFO | train rank=0 | step=76680/100000 progress=76.7% loss=0.046715 lr=1.414551e-06 elapsed=28:28:14 avg_step=1.34s recent_step=1.43s eta=08:39:30 +2026-08-26 08:23:56 | INFO | train rank=0 | step=76690/100000 progress=76.7% loss=0.029160 lr=1.413398e-06 elapsed=28:28:26 avg_step=1.34s recent_step=1.21s eta=08:39:16 +2026-08-26 08:24:07 | INFO | train rank=0 | step=76700/100000 progress=76.7% loss=0.046555 lr=1.412246e-06 elapsed=28:28:37 avg_step=1.34s recent_step=1.15s eta=08:39:02 +2026-08-26 08:24:20 | INFO | train rank=0 | step=76710/100000 progress=76.7% loss=0.035217 lr=1.411095e-06 elapsed=28:28:50 avg_step=1.34s recent_step=1.30s eta=08:38:49 +2026-08-26 08:24:35 | INFO | train rank=0 | step=76720/100000 progress=76.7% loss=0.045380 lr=1.409944e-06 elapsed=28:29:05 avg_step=1.34s recent_step=1.49s eta=08:38:36 +2026-08-26 08:24:47 | INFO | train rank=0 | step=76730/100000 progress=76.7% loss=0.039032 lr=1.408793e-06 elapsed=28:29:17 avg_step=1.34s recent_step=1.17s eta=08:38:22 +2026-08-26 08:25:01 | INFO | train rank=0 | step=76740/100000 progress=76.7% loss=0.044205 lr=1.407643e-06 elapsed=28:29:31 avg_step=1.34s recent_step=1.39s eta=08:38:09 +2026-08-26 08:25:18 | INFO | train rank=0 | step=76750/100000 progress=76.8% loss=0.030525 lr=1.406493e-06 elapsed=28:29:48 avg_step=1.34s recent_step=1.70s eta=08:37:57 +2026-08-26 08:25:32 | INFO | train rank=0 | step=76760/100000 progress=76.8% loss=0.038162 lr=1.405344e-06 elapsed=28:30:02 avg_step=1.34s recent_step=1.43s eta=08:37:44 +2026-08-26 08:25:44 | INFO | train rank=0 | step=76770/100000 progress=76.8% loss=0.030350 lr=1.404194e-06 elapsed=28:30:14 avg_step=1.34s recent_step=1.23s eta=08:37:30 +2026-08-26 08:26:00 | INFO | train rank=0 | step=76780/100000 progress=76.8% loss=0.038567 lr=1.403046e-06 elapsed=28:30:30 avg_step=1.34s recent_step=1.57s eta=08:37:17 +2026-08-26 08:26:12 | INFO | train rank=0 | step=76790/100000 progress=76.8% loss=0.048111 lr=1.401897e-06 elapsed=28:30:42 avg_step=1.34s recent_step=1.16s eta=08:37:03 +2026-08-26 08:26:24 | INFO | train rank=0 | step=76800/100000 progress=76.8% loss=0.061066 lr=1.400750e-06 elapsed=28:30:54 avg_step=1.34s recent_step=1.24s eta=08:36:50 +2026-08-26 08:26:41 | INFO | train rank=0 | step=76810/100000 progress=76.8% loss=0.039925 lr=1.399602e-06 elapsed=28:31:11 avg_step=1.34s recent_step=1.71s eta=08:36:37 +2026-08-26 08:26:53 | INFO | train rank=0 | step=76820/100000 progress=76.8% loss=0.041122 lr=1.398455e-06 elapsed=28:31:23 avg_step=1.34s recent_step=1.17s eta=08:36:24 +2026-08-26 08:27:07 | INFO | train rank=0 | step=76830/100000 progress=76.8% loss=0.059509 lr=1.397308e-06 elapsed=28:31:37 avg_step=1.34s recent_step=1.43s eta=08:36:11 +2026-08-26 08:27:26 | INFO | train rank=0 | step=76840/100000 progress=76.8% loss=0.035294 lr=1.396162e-06 elapsed=28:31:56 avg_step=1.34s recent_step=1.84s eta=08:35:59 +2026-08-26 08:27:37 | INFO | train rank=0 | step=76850/100000 progress=76.8% loss=0.035881 lr=1.395016e-06 elapsed=28:32:07 avg_step=1.34s recent_step=1.18s eta=08:35:45 +2026-08-26 08:27:49 | INFO | train rank=0 | step=76860/100000 progress=76.9% loss=0.077606 lr=1.393870e-06 elapsed=28:32:18 avg_step=1.34s recent_step=1.11s eta=08:35:31 +2026-08-26 08:28:00 | INFO | train rank=0 | step=76870/100000 progress=76.9% loss=0.041428 lr=1.392725e-06 elapsed=28:32:30 avg_step=1.34s recent_step=1.18s eta=08:35:17 +2026-08-26 08:28:16 | INFO | train rank=0 | step=76880/100000 progress=76.9% loss=0.044685 lr=1.391580e-06 elapsed=28:32:46 avg_step=1.34s recent_step=1.56s eta=08:35:04 +2026-08-26 08:28:31 | INFO | train rank=0 | step=76890/100000 progress=76.9% loss=0.033577 lr=1.390436e-06 elapsed=28:33:01 avg_step=1.34s recent_step=1.47s eta=08:34:51 +2026-08-26 08:28:43 | INFO | train rank=0 | step=76900/100000 progress=76.9% loss=0.062805 lr=1.389292e-06 elapsed=28:33:13 avg_step=1.34s recent_step=1.23s eta=08:34:38 +2026-08-26 08:28:58 | INFO | train rank=0 | step=76910/100000 progress=76.9% loss=0.080048 lr=1.388148e-06 elapsed=28:33:28 avg_step=1.34s recent_step=1.46s eta=08:34:25 +2026-08-26 08:29:10 | INFO | train rank=0 | step=76920/100000 progress=76.9% loss=0.033760 lr=1.387005e-06 elapsed=28:33:40 avg_step=1.34s recent_step=1.27s eta=08:34:11 +2026-08-26 08:29:25 | INFO | train rank=0 | step=76930/100000 progress=76.9% loss=0.050781 lr=1.385862e-06 elapsed=28:33:55 avg_step=1.34s recent_step=1.50s eta=08:33:58 +2026-08-26 08:29:37 | INFO | train rank=0 | step=76940/100000 progress=76.9% loss=0.042496 lr=1.384720e-06 elapsed=28:34:06 avg_step=1.34s recent_step=1.13s eta=08:33:44 +2026-08-26 08:29:52 | INFO | train rank=0 | step=76950/100000 progress=77.0% loss=0.060822 lr=1.383578e-06 elapsed=28:34:22 avg_step=1.34s recent_step=1.52s eta=08:33:31 +2026-08-26 08:30:06 | INFO | train rank=0 | step=76960/100000 progress=77.0% loss=0.046036 lr=1.382436e-06 elapsed=28:34:36 avg_step=1.34s recent_step=1.41s eta=08:33:18 +2026-08-26 08:30:18 | INFO | train rank=0 | step=76970/100000 progress=77.0% loss=0.025124 lr=1.381295e-06 elapsed=28:34:48 avg_step=1.34s recent_step=1.19s eta=08:33:04 +2026-08-26 08:30:29 | INFO | train rank=0 | step=76980/100000 progress=77.0% loss=0.056519 lr=1.380154e-06 elapsed=28:34:59 avg_step=1.34s recent_step=1.13s eta=08:32:50 +2026-08-26 08:30:40 | INFO | train rank=0 | step=76990/100000 progress=77.0% loss=0.042618 lr=1.379014e-06 elapsed=28:35:10 avg_step=1.34s recent_step=1.13s eta=08:32:36 +2026-08-26 08:30:56 | INFO | train rank=0 | step=77000/100000 progress=77.0% loss=0.040817 lr=1.377874e-06 elapsed=28:35:26 avg_step=1.34s recent_step=1.58s eta=08:32:24 +2026-08-26 08:31:08 | INFO | train rank=0 | step=77010/100000 progress=77.0% loss=0.085144 lr=1.376734e-06 elapsed=28:35:38 avg_step=1.34s recent_step=1.21s eta=08:32:10 +2026-08-26 08:31:22 | INFO | train rank=0 | step=77020/100000 progress=77.0% loss=0.074432 lr=1.375595e-06 elapsed=28:35:52 avg_step=1.34s recent_step=1.40s eta=08:31:57 +2026-08-26 08:31:35 | INFO | train rank=0 | step=77030/100000 progress=77.0% loss=0.028687 lr=1.374456e-06 elapsed=28:36:05 avg_step=1.34s recent_step=1.26s eta=08:31:43 +2026-08-26 08:31:49 | INFO | train rank=0 | step=77040/100000 progress=77.0% loss=0.075439 lr=1.373318e-06 elapsed=28:36:19 avg_step=1.34s recent_step=1.44s eta=08:31:30 +2026-08-26 08:32:03 | INFO | train rank=0 | step=77050/100000 progress=77.0% loss=0.051514 lr=1.372180e-06 elapsed=28:36:33 avg_step=1.34s recent_step=1.42s eta=08:31:17 +2026-08-26 08:32:18 | INFO | train rank=0 | step=77060/100000 progress=77.1% loss=0.049240 lr=1.371042e-06 elapsed=28:36:48 avg_step=1.34s recent_step=1.46s eta=08:31:04 +2026-08-26 08:32:29 | INFO | train rank=0 | step=77070/100000 progress=77.1% loss=0.056412 lr=1.369905e-06 elapsed=28:36:59 avg_step=1.34s recent_step=1.10s eta=08:30:50 +2026-08-26 08:32:41 | INFO | train rank=0 | step=77080/100000 progress=77.1% loss=0.060715 lr=1.368768e-06 elapsed=28:37:11 avg_step=1.34s recent_step=1.18s eta=08:30:36 +2026-08-26 08:32:55 | INFO | train rank=0 | step=77090/100000 progress=77.1% loss=0.094238 lr=1.367632e-06 elapsed=28:37:25 avg_step=1.34s recent_step=1.44s eta=08:30:23 +2026-08-26 08:33:09 | INFO | train rank=0 | step=77100/100000 progress=77.1% loss=0.056267 lr=1.366496e-06 elapsed=28:37:39 avg_step=1.34s recent_step=1.39s eta=08:30:10 +2026-08-26 08:33:21 | INFO | train rank=0 | step=77110/100000 progress=77.1% loss=0.029114 lr=1.365360e-06 elapsed=28:37:51 avg_step=1.34s recent_step=1.16s eta=08:29:56 +2026-08-26 08:33:35 | INFO | train rank=0 | step=77120/100000 progress=77.1% loss=0.038162 lr=1.364225e-06 elapsed=28:38:05 avg_step=1.34s recent_step=1.41s eta=08:29:43 +2026-08-26 08:33:54 | INFO | train rank=0 | step=77130/100000 progress=77.1% loss=0.034912 lr=1.363090e-06 elapsed=28:38:24 avg_step=1.34s recent_step=1.91s eta=08:29:31 +2026-08-26 08:34:06 | INFO | train rank=0 | step=77140/100000 progress=77.1% loss=0.042480 lr=1.361955e-06 elapsed=28:38:36 avg_step=1.34s recent_step=1.21s eta=08:29:17 +2026-08-26 08:34:18 | INFO | train rank=0 | step=77150/100000 progress=77.1% loss=0.046860 lr=1.360821e-06 elapsed=28:38:47 avg_step=1.34s recent_step=1.17s eta=08:29:04 +2026-08-26 08:34:31 | INFO | train rank=0 | step=77160/100000 progress=77.2% loss=0.039307 lr=1.359688e-06 elapsed=28:39:01 avg_step=1.34s recent_step=1.39s eta=08:28:50 +2026-08-26 08:34:48 | INFO | train rank=0 | step=77170/100000 progress=77.2% loss=0.051636 lr=1.358554e-06 elapsed=28:39:18 avg_step=1.34s recent_step=1.63s eta=08:28:38 +2026-08-26 08:35:00 | INFO | train rank=0 | step=77180/100000 progress=77.2% loss=0.100342 lr=1.357421e-06 elapsed=28:39:29 avg_step=1.34s recent_step=1.18s eta=08:28:24 +2026-08-26 08:35:16 | INFO | train rank=0 | step=77190/100000 progress=77.2% loss=0.030701 lr=1.356289e-06 elapsed=28:39:46 avg_step=1.34s recent_step=1.68s eta=08:28:12 +2026-08-26 08:35:28 | INFO | train rank=0 | step=77200/100000 progress=77.2% loss=0.034225 lr=1.355157e-06 elapsed=28:39:58 avg_step=1.34s recent_step=1.14s eta=08:27:58 +2026-08-26 08:35:47 | INFO | train rank=0 | step=77210/100000 progress=77.2% loss=0.041107 lr=1.354025e-06 elapsed=28:40:17 avg_step=1.34s recent_step=1.90s eta=08:27:46 +2026-08-26 08:35:58 | INFO | train rank=0 | step=77220/100000 progress=77.2% loss=0.064270 lr=1.352894e-06 elapsed=28:40:28 avg_step=1.34s recent_step=1.11s eta=08:27:32 +2026-08-26 08:36:10 | INFO | train rank=0 | step=77230/100000 progress=77.2% loss=0.032791 lr=1.351763e-06 elapsed=28:40:39 avg_step=1.34s recent_step=1.17s eta=08:27:18 +2026-08-26 08:36:23 | INFO | train rank=0 | step=77240/100000 progress=77.2% loss=0.031990 lr=1.350633e-06 elapsed=28:40:53 avg_step=1.34s recent_step=1.33s eta=08:27:05 +2026-08-26 08:36:38 | INFO | train rank=0 | step=77250/100000 progress=77.2% loss=0.037727 lr=1.349502e-06 elapsed=28:41:08 avg_step=1.34s recent_step=1.55s eta=08:26:52 +2026-08-26 08:36:53 | INFO | train rank=0 | step=77260/100000 progress=77.3% loss=0.079803 lr=1.348373e-06 elapsed=28:41:23 avg_step=1.34s recent_step=1.46s eta=08:26:39 +2026-08-26 08:37:04 | INFO | train rank=0 | step=77270/100000 progress=77.3% loss=0.043724 lr=1.347243e-06 elapsed=28:41:34 avg_step=1.34s recent_step=1.13s eta=08:26:25 +2026-08-26 08:37:16 | INFO | train rank=0 | step=77280/100000 progress=77.3% loss=0.042099 lr=1.346115e-06 elapsed=28:41:46 avg_step=1.34s recent_step=1.15s eta=08:26:11 +2026-08-26 08:37:33 | INFO | train rank=0 | step=77290/100000 progress=77.3% loss=0.042862 lr=1.344986e-06 elapsed=28:42:03 avg_step=1.34s recent_step=1.75s eta=08:25:59 +2026-08-26 08:37:45 | INFO | train rank=0 | step=77300/100000 progress=77.3% loss=0.055527 lr=1.343858e-06 elapsed=28:42:15 avg_step=1.34s recent_step=1.17s eta=08:25:45 +2026-08-26 08:37:58 | INFO | train rank=0 | step=77310/100000 progress=77.3% loss=0.050934 lr=1.342730e-06 elapsed=28:42:28 avg_step=1.34s recent_step=1.29s eta=08:25:32 +2026-08-26 08:38:10 | INFO | train rank=0 | step=77320/100000 progress=77.3% loss=0.035370 lr=1.341603e-06 elapsed=28:42:40 avg_step=1.34s recent_step=1.18s eta=08:25:18 +2026-08-26 08:38:21 | INFO | train rank=0 | step=77330/100000 progress=77.3% loss=0.033524 lr=1.340476e-06 elapsed=28:42:51 avg_step=1.34s recent_step=1.13s eta=08:25:04 +2026-08-26 08:38:35 | INFO | train rank=0 | step=77340/100000 progress=77.3% loss=0.052856 lr=1.339350e-06 elapsed=28:43:05 avg_step=1.34s recent_step=1.41s eta=08:24:51 +2026-08-26 08:38:48 | INFO | train rank=0 | step=77350/100000 progress=77.3% loss=0.035591 lr=1.338224e-06 elapsed=28:43:17 avg_step=1.34s recent_step=1.24s eta=08:24:37 +2026-08-26 08:39:02 | INFO | train rank=0 | step=77360/100000 progress=77.4% loss=0.041779 lr=1.337098e-06 elapsed=28:43:32 avg_step=1.34s recent_step=1.42s eta=08:24:24 +2026-08-26 08:39:15 | INFO | train rank=0 | step=77370/100000 progress=77.4% loss=0.052505 lr=1.335973e-06 elapsed=28:43:45 avg_step=1.34s recent_step=1.35s eta=08:24:11 +2026-08-26 08:39:27 | INFO | train rank=0 | step=77380/100000 progress=77.4% loss=0.059601 lr=1.334848e-06 elapsed=28:43:57 avg_step=1.34s recent_step=1.21s eta=08:23:57 +2026-08-26 08:39:39 | INFO | train rank=0 | step=77390/100000 progress=77.4% loss=0.049545 lr=1.333723e-06 elapsed=28:44:09 avg_step=1.34s recent_step=1.15s eta=08:23:43 +2026-08-26 08:39:54 | INFO | train rank=0 | step=77400/100000 progress=77.4% loss=0.044403 lr=1.332599e-06 elapsed=28:44:23 avg_step=1.34s recent_step=1.48s eta=08:23:30 +2026-08-26 08:40:07 | INFO | train rank=0 | step=77410/100000 progress=77.4% loss=0.043777 lr=1.331476e-06 elapsed=28:44:37 avg_step=1.34s recent_step=1.31s eta=08:23:16 +2026-08-26 08:40:21 | INFO | train rank=0 | step=77420/100000 progress=77.4% loss=0.029236 lr=1.330352e-06 elapsed=28:44:51 avg_step=1.34s recent_step=1.47s eta=08:23:03 +2026-08-26 08:40:35 | INFO | train rank=0 | step=77430/100000 progress=77.4% loss=0.057953 lr=1.329229e-06 elapsed=28:45:05 avg_step=1.34s recent_step=1.40s eta=08:22:50 +2026-08-26 08:40:47 | INFO | train rank=0 | step=77440/100000 progress=77.4% loss=0.100708 lr=1.328107e-06 elapsed=28:45:17 avg_step=1.34s recent_step=1.14s eta=08:22:36 +2026-08-26 08:40:59 | INFO | train rank=0 | step=77450/100000 progress=77.5% loss=0.025688 lr=1.326985e-06 elapsed=28:45:29 avg_step=1.34s recent_step=1.27s eta=08:22:23 +2026-08-26 08:41:15 | INFO | train rank=0 | step=77460/100000 progress=77.5% loss=0.067184 lr=1.325863e-06 elapsed=28:45:45 avg_step=1.34s recent_step=1.52s eta=08:22:10 +2026-08-26 08:41:28 | INFO | train rank=0 | step=77470/100000 progress=77.5% loss=0.044479 lr=1.324742e-06 elapsed=28:45:58 avg_step=1.34s recent_step=1.37s eta=08:21:57 +2026-08-26 08:41:43 | INFO | train rank=0 | step=77480/100000 progress=77.5% loss=0.083282 lr=1.323621e-06 elapsed=28:46:13 avg_step=1.34s recent_step=1.52s eta=08:21:44 +2026-08-26 08:41:59 | INFO | train rank=0 | step=77490/100000 progress=77.5% loss=0.053833 lr=1.322501e-06 elapsed=28:46:29 avg_step=1.34s recent_step=1.60s eta=08:21:31 +2026-08-26 08:42:11 | INFO | train rank=0 | step=77500/100000 progress=77.5% loss=0.044044 lr=1.321380e-06 elapsed=28:46:41 avg_step=1.34s recent_step=1.16s eta=08:21:17 +2026-08-26 08:42:23 | INFO | train rank=0 | step=77510/100000 progress=77.5% loss=0.043274 lr=1.320261e-06 elapsed=28:46:53 avg_step=1.34s recent_step=1.23s eta=08:21:04 +2026-08-26 08:42:41 | INFO | train rank=0 | step=77520/100000 progress=77.5% loss=0.037766 lr=1.319142e-06 elapsed=28:47:11 avg_step=1.34s recent_step=1.73s eta=08:20:51 +2026-08-26 08:42:53 | INFO | train rank=0 | step=77530/100000 progress=77.5% loss=0.039078 lr=1.318023e-06 elapsed=28:47:23 avg_step=1.34s recent_step=1.20s eta=08:20:38 +2026-08-26 08:43:07 | INFO | train rank=0 | step=77540/100000 progress=77.5% loss=0.029716 lr=1.316904e-06 elapsed=28:47:37 avg_step=1.34s recent_step=1.45s eta=08:20:25 +2026-08-26 08:43:19 | INFO | train rank=0 | step=77550/100000 progress=77.5% loss=0.045990 lr=1.315786e-06 elapsed=28:47:49 avg_step=1.34s recent_step=1.15s eta=08:20:11 +2026-08-26 08:43:31 | INFO | train rank=0 | step=77560/100000 progress=77.6% loss=0.035591 lr=1.314669e-06 elapsed=28:48:01 avg_step=1.34s recent_step=1.21s eta=08:19:57 +2026-08-26 08:43:46 | INFO | train rank=0 | step=77570/100000 progress=77.6% loss=0.045189 lr=1.313551e-06 elapsed=28:48:16 avg_step=1.34s recent_step=1.57s eta=08:19:44 +2026-08-26 08:43:58 | INFO | train rank=0 | step=77580/100000 progress=77.6% loss=0.048141 lr=1.312434e-06 elapsed=28:48:28 avg_step=1.34s recent_step=1.21s eta=08:19:31 +2026-08-26 08:44:10 | INFO | train rank=0 | step=77590/100000 progress=77.6% loss=0.043213 lr=1.311318e-06 elapsed=28:48:40 avg_step=1.34s recent_step=1.14s eta=08:19:17 +2026-08-26 08:44:28 | INFO | train rank=0 | step=77600/100000 progress=77.6% loss=0.036453 lr=1.310202e-06 elapsed=28:48:58 avg_step=1.34s recent_step=1.84s eta=08:19:05 +2026-08-26 08:44:43 | INFO | train rank=0 | step=77610/100000 progress=77.6% loss=0.027775 lr=1.309086e-06 elapsed=28:49:13 avg_step=1.34s recent_step=1.50s eta=08:18:52 +2026-08-26 08:44:54 | INFO | train rank=0 | step=77620/100000 progress=77.6% loss=0.033737 lr=1.307971e-06 elapsed=28:49:24 avg_step=1.34s recent_step=1.10s eta=08:18:38 +2026-08-26 08:45:07 | INFO | train rank=0 | step=77630/100000 progress=77.6% loss=0.043869 lr=1.306856e-06 elapsed=28:49:37 avg_step=1.34s recent_step=1.26s eta=08:18:24 +2026-08-26 08:45:21 | INFO | train rank=0 | step=77640/100000 progress=77.6% loss=0.027145 lr=1.305742e-06 elapsed=28:49:51 avg_step=1.34s recent_step=1.45s eta=08:18:11 +2026-08-26 08:45:33 | INFO | train rank=0 | step=77650/100000 progress=77.6% loss=0.038208 lr=1.304628e-06 elapsed=28:50:03 avg_step=1.34s recent_step=1.15s eta=08:17:57 +2026-08-26 08:45:44 | INFO | train rank=0 | step=77660/100000 progress=77.7% loss=0.037720 lr=1.303514e-06 elapsed=28:50:14 avg_step=1.34s recent_step=1.14s eta=08:17:43 +2026-08-26 08:45:59 | INFO | train rank=0 | step=77670/100000 progress=77.7% loss=0.073853 lr=1.302401e-06 elapsed=28:50:29 avg_step=1.34s recent_step=1.45s eta=08:17:30 +2026-08-26 08:46:14 | INFO | train rank=0 | step=77680/100000 progress=77.7% loss=0.053024 lr=1.301288e-06 elapsed=28:50:44 avg_step=1.34s recent_step=1.52s eta=08:17:17 +2026-08-26 08:46:31 | INFO | train rank=0 | step=77690/100000 progress=77.7% loss=0.032295 lr=1.300176e-06 elapsed=28:51:01 avg_step=1.34s recent_step=1.67s eta=08:17:05 +2026-08-26 08:46:42 | INFO | train rank=0 | step=77700/100000 progress=77.7% loss=0.047760 lr=1.299064e-06 elapsed=28:51:12 avg_step=1.34s recent_step=1.16s eta=08:16:51 +2026-08-26 08:46:54 | INFO | train rank=0 | step=77710/100000 progress=77.7% loss=0.032570 lr=1.297952e-06 elapsed=28:51:24 avg_step=1.34s recent_step=1.20s eta=08:16:37 +2026-08-26 08:47:06 | INFO | train rank=0 | step=77720/100000 progress=77.7% loss=0.063751 lr=1.296841e-06 elapsed=28:51:36 avg_step=1.34s recent_step=1.15s eta=08:16:23 +2026-08-26 08:47:20 | INFO | train rank=0 | step=77730/100000 progress=77.7% loss=0.050873 lr=1.295730e-06 elapsed=28:51:50 avg_step=1.34s recent_step=1.46s eta=08:16:10 +2026-08-26 08:47:39 | INFO | train rank=0 | step=77740/100000 progress=77.7% loss=0.042381 lr=1.294620e-06 elapsed=28:52:09 avg_step=1.34s recent_step=1.89s eta=08:15:59 +2026-08-26 08:47:50 | INFO | train rank=0 | step=77750/100000 progress=77.8% loss=0.032143 lr=1.293510e-06 elapsed=28:52:20 avg_step=1.34s recent_step=1.10s eta=08:15:45 +2026-08-26 08:48:05 | INFO | train rank=0 | step=77760/100000 progress=77.8% loss=0.043327 lr=1.292400e-06 elapsed=28:52:35 avg_step=1.34s recent_step=1.45s eta=08:15:32 +2026-08-26 08:48:16 | INFO | train rank=0 | step=77770/100000 progress=77.8% loss=0.034515 lr=1.291291e-06 elapsed=28:52:46 avg_step=1.34s recent_step=1.16s eta=08:15:18 +2026-08-26 08:48:31 | INFO | train rank=0 | step=77780/100000 progress=77.8% loss=0.037277 lr=1.290182e-06 elapsed=28:53:01 avg_step=1.34s recent_step=1.46s eta=08:15:05 +2026-08-26 08:48:48 | INFO | train rank=0 | step=77790/100000 progress=77.8% loss=0.027161 lr=1.289074e-06 elapsed=28:53:18 avg_step=1.34s recent_step=1.68s eta=08:14:52 +2026-08-26 08:49:01 | INFO | train rank=0 | step=77800/100000 progress=77.8% loss=0.043243 lr=1.287966e-06 elapsed=28:53:31 avg_step=1.34s recent_step=1.29s eta=08:14:39 +2026-08-26 08:49:12 | INFO | train rank=0 | step=77810/100000 progress=77.8% loss=0.025185 lr=1.286859e-06 elapsed=28:53:42 avg_step=1.34s recent_step=1.17s eta=08:14:25 +2026-08-26 08:49:27 | INFO | train rank=0 | step=77820/100000 progress=77.8% loss=0.048119 lr=1.285751e-06 elapsed=28:53:57 avg_step=1.34s recent_step=1.43s eta=08:14:12 +2026-08-26 08:49:43 | INFO | train rank=0 | step=77830/100000 progress=77.8% loss=0.041534 lr=1.284645e-06 elapsed=28:54:13 avg_step=1.34s recent_step=1.65s eta=08:13:59 +2026-08-26 08:49:55 | INFO | train rank=0 | step=77840/100000 progress=77.8% loss=0.040222 lr=1.283538e-06 elapsed=28:54:25 avg_step=1.34s recent_step=1.18s eta=08:13:46 +2026-08-26 08:50:10 | INFO | train rank=0 | step=77850/100000 progress=77.8% loss=0.061066 lr=1.282432e-06 elapsed=28:54:40 avg_step=1.34s recent_step=1.47s eta=08:13:33 +2026-08-26 08:50:21 | INFO | train rank=0 | step=77860/100000 progress=77.9% loss=0.071716 lr=1.281327e-06 elapsed=28:54:51 avg_step=1.34s recent_step=1.10s eta=08:13:19 +2026-08-26 08:50:34 | INFO | train rank=0 | step=77870/100000 progress=77.9% loss=0.059387 lr=1.280222e-06 elapsed=28:55:04 avg_step=1.34s recent_step=1.35s eta=08:13:05 +2026-08-26 08:50:52 | INFO | train rank=0 | step=77880/100000 progress=77.9% loss=0.056839 lr=1.279117e-06 elapsed=28:55:22 avg_step=1.34s recent_step=1.76s eta=08:12:53 +2026-08-26 08:51:03 | INFO | train rank=0 | step=77890/100000 progress=77.9% loss=0.087616 lr=1.278013e-06 elapsed=28:55:33 avg_step=1.34s recent_step=1.13s eta=08:12:39 +2026-08-26 08:51:15 | INFO | train rank=0 | step=77900/100000 progress=77.9% loss=0.067780 lr=1.276909e-06 elapsed=28:55:45 avg_step=1.34s recent_step=1.17s eta=08:12:25 +2026-08-26 08:51:26 | INFO | train rank=0 | step=77910/100000 progress=77.9% loss=0.041672 lr=1.275806e-06 elapsed=28:55:56 avg_step=1.34s recent_step=1.14s eta=08:12:11 +2026-08-26 08:51:44 | INFO | train rank=0 | step=77920/100000 progress=77.9% loss=0.037628 lr=1.274702e-06 elapsed=28:56:14 avg_step=1.34s recent_step=1.76s eta=08:11:59 +2026-08-26 08:51:56 | INFO | train rank=0 | step=77930/100000 progress=77.9% loss=0.081421 lr=1.273600e-06 elapsed=28:56:26 avg_step=1.34s recent_step=1.18s eta=08:11:45 +2026-08-26 08:52:08 | INFO | train rank=0 | step=77940/100000 progress=77.9% loss=0.038895 lr=1.272498e-06 elapsed=28:56:38 avg_step=1.34s recent_step=1.18s eta=08:11:32 +2026-08-26 08:52:19 | INFO | train rank=0 | step=77950/100000 progress=78.0% loss=0.037041 lr=1.271396e-06 elapsed=28:56:49 avg_step=1.34s recent_step=1.17s eta=08:11:18 +2026-08-26 08:52:35 | INFO | train rank=0 | step=77960/100000 progress=78.0% loss=0.046671 lr=1.270294e-06 elapsed=28:57:05 avg_step=1.34s recent_step=1.56s eta=08:11:05 +2026-08-26 08:52:49 | INFO | train rank=0 | step=77970/100000 progress=78.0% loss=0.035950 lr=1.269193e-06 elapsed=28:57:19 avg_step=1.34s recent_step=1.42s eta=08:10:52 +2026-08-26 08:53:02 | INFO | train rank=0 | step=77980/100000 progress=78.0% loss=0.037659 lr=1.268093e-06 elapsed=28:57:32 avg_step=1.34s recent_step=1.29s eta=08:10:38 +2026-08-26 08:53:18 | INFO | train rank=0 | step=77990/100000 progress=78.0% loss=0.051071 lr=1.266992e-06 elapsed=28:57:48 avg_step=1.34s recent_step=1.57s eta=08:10:26 +2026-08-26 08:53:31 | INFO | train rank=0 | step=78000/100000 progress=78.0% loss=0.050308 lr=1.265893e-06 elapsed=28:58:01 avg_step=1.34s recent_step=1.38s eta=08:10:12 +2026-08-26 08:53:43 | INFO | train rank=0 | step=78010/100000 progress=78.0% loss=0.060623 lr=1.264793e-06 elapsed=28:58:13 avg_step=1.34s recent_step=1.16s eta=08:09:58 +2026-08-26 08:53:55 | INFO | train rank=0 | step=78020/100000 progress=78.0% loss=0.071442 lr=1.263694e-06 elapsed=28:58:25 avg_step=1.34s recent_step=1.18s eta=08:09:45 +2026-08-26 08:54:07 | INFO | train rank=0 | step=78030/100000 progress=78.0% loss=0.035385 lr=1.262596e-06 elapsed=28:58:36 avg_step=1.34s recent_step=1.18s eta=08:09:31 +2026-08-26 08:54:21 | INFO | train rank=0 | step=78040/100000 progress=78.0% loss=0.035744 lr=1.261497e-06 elapsed=28:58:51 avg_step=1.34s recent_step=1.47s eta=08:09:18 +2026-08-26 08:54:34 | INFO | train rank=0 | step=78050/100000 progress=78.0% loss=0.039017 lr=1.260400e-06 elapsed=28:59:04 avg_step=1.34s recent_step=1.23s eta=08:09:04 +2026-08-26 08:54:48 | INFO | train rank=0 | step=78060/100000 progress=78.1% loss=0.110046 lr=1.259302e-06 elapsed=28:59:18 avg_step=1.34s recent_step=1.42s eta=08:08:51 +2026-08-26 08:55:03 | INFO | train rank=0 | step=78070/100000 progress=78.1% loss=0.048859 lr=1.258205e-06 elapsed=28:59:33 avg_step=1.34s recent_step=1.56s eta=08:08:38 +2026-08-26 08:55:15 | INFO | train rank=0 | step=78080/100000 progress=78.1% loss=0.049232 lr=1.257109e-06 elapsed=28:59:45 avg_step=1.34s recent_step=1.16s eta=08:08:24 +2026-08-26 08:55:29 | INFO | train rank=0 | step=78090/100000 progress=78.1% loss=0.034637 lr=1.256013e-06 elapsed=28:59:59 avg_step=1.34s recent_step=1.39s eta=08:08:11 +2026-08-26 08:55:45 | INFO | train rank=0 | step=78100/100000 progress=78.1% loss=0.059204 lr=1.254917e-06 elapsed=29:00:15 avg_step=1.34s recent_step=1.65s eta=08:07:59 +2026-08-26 08:55:57 | INFO | train rank=0 | step=78110/100000 progress=78.1% loss=0.039520 lr=1.253822e-06 elapsed=29:00:27 avg_step=1.34s recent_step=1.16s eta=08:07:45 +2026-08-26 08:56:14 | INFO | train rank=0 | step=78120/100000 progress=78.1% loss=0.078217 lr=1.252727e-06 elapsed=29:00:43 avg_step=1.34s recent_step=1.66s eta=08:07:32 +2026-08-26 08:56:25 | INFO | train rank=0 | step=78130/100000 progress=78.1% loss=0.035423 lr=1.251632e-06 elapsed=29:00:55 avg_step=1.34s recent_step=1.14s eta=08:07:18 +2026-08-26 08:56:40 | INFO | train rank=0 | step=78140/100000 progress=78.1% loss=0.051369 lr=1.250538e-06 elapsed=29:01:10 avg_step=1.34s recent_step=1.54s eta=08:07:06 +2026-08-26 08:56:53 | INFO | train rank=0 | step=78150/100000 progress=78.1% loss=0.057831 lr=1.249445e-06 elapsed=29:01:23 avg_step=1.34s recent_step=1.31s eta=08:06:52 +2026-08-26 08:57:08 | INFO | train rank=0 | step=78160/100000 progress=78.2% loss=0.040367 lr=1.248351e-06 elapsed=29:01:37 avg_step=1.34s recent_step=1.42s eta=08:06:39 +2026-08-26 08:57:19 | INFO | train rank=0 | step=78170/100000 progress=78.2% loss=0.053955 lr=1.247259e-06 elapsed=29:01:49 avg_step=1.34s recent_step=1.17s eta=08:06:25 +2026-08-26 08:57:30 | INFO | train rank=0 | step=78180/100000 progress=78.2% loss=0.099731 lr=1.246166e-06 elapsed=29:02:00 avg_step=1.34s recent_step=1.12s eta=08:06:11 +2026-08-26 08:57:47 | INFO | train rank=0 | step=78190/100000 progress=78.2% loss=0.053177 lr=1.245074e-06 elapsed=29:02:17 avg_step=1.34s recent_step=1.64s eta=08:05:59 +2026-08-26 08:58:02 | INFO | train rank=0 | step=78200/100000 progress=78.2% loss=0.043655 lr=1.243982e-06 elapsed=29:02:32 avg_step=1.34s recent_step=1.47s eta=08:05:46 +2026-08-26 08:58:13 | INFO | train rank=0 | step=78210/100000 progress=78.2% loss=0.061432 lr=1.242891e-06 elapsed=29:02:43 avg_step=1.34s recent_step=1.17s eta=08:05:32 +2026-08-26 08:58:25 | INFO | train rank=0 | step=78220/100000 progress=78.2% loss=0.032547 lr=1.241800e-06 elapsed=29:02:55 avg_step=1.34s recent_step=1.18s eta=08:05:18 +2026-08-26 08:58:40 | INFO | train rank=0 | step=78230/100000 progress=78.2% loss=0.032654 lr=1.240710e-06 elapsed=29:03:10 avg_step=1.34s recent_step=1.49s eta=08:05:05 +2026-08-26 08:58:54 | INFO | train rank=0 | step=78240/100000 progress=78.2% loss=0.105896 lr=1.239620e-06 elapsed=29:03:24 avg_step=1.34s recent_step=1.43s eta=08:04:52 +2026-08-26 08:59:09 | INFO | train rank=0 | step=78250/100000 progress=78.2% loss=0.046257 lr=1.238531e-06 elapsed=29:03:39 avg_step=1.34s recent_step=1.47s eta=08:04:39 +2026-08-26 08:59:20 | INFO | train rank=0 | step=78260/100000 progress=78.3% loss=0.031754 lr=1.237441e-06 elapsed=29:03:50 avg_step=1.34s recent_step=1.11s eta=08:04:25 +2026-08-26 08:59:32 | INFO | train rank=0 | step=78270/100000 progress=78.3% loss=0.024002 lr=1.236353e-06 elapsed=29:04:01 avg_step=1.34s recent_step=1.15s eta=08:04:11 +2026-08-26 08:59:52 | INFO | train rank=0 | step=78280/100000 progress=78.3% loss=0.036079 lr=1.235264e-06 elapsed=29:04:21 avg_step=1.34s recent_step=2.00s eta=08:04:00 +2026-08-26 09:00:05 | INFO | train rank=0 | step=78290/100000 progress=78.3% loss=0.057709 lr=1.234176e-06 elapsed=29:04:35 avg_step=1.34s recent_step=1.34s eta=08:03:46 +2026-08-26 09:00:16 | INFO | train rank=0 | step=78300/100000 progress=78.3% loss=0.072845 lr=1.233089e-06 elapsed=29:04:46 avg_step=1.34s recent_step=1.14s eta=08:03:32 +2026-08-26 09:00:28 | INFO | train rank=0 | step=78310/100000 progress=78.3% loss=0.033737 lr=1.232002e-06 elapsed=29:04:58 avg_step=1.34s recent_step=1.14s eta=08:03:18 +2026-08-26 09:00:41 | INFO | train rank=0 | step=78320/100000 progress=78.3% loss=0.030121 lr=1.230915e-06 elapsed=29:05:11 avg_step=1.34s recent_step=1.31s eta=08:03:05 +2026-08-26 09:00:53 | INFO | train rank=0 | step=78330/100000 progress=78.3% loss=0.070831 lr=1.229829e-06 elapsed=29:05:23 avg_step=1.34s recent_step=1.21s eta=08:02:51 +2026-08-26 09:01:05 | INFO | train rank=0 | step=78340/100000 progress=78.3% loss=0.053314 lr=1.228743e-06 elapsed=29:05:35 avg_step=1.34s recent_step=1.16s eta=08:02:37 +2026-08-26 09:01:16 | INFO | train rank=0 | step=78350/100000 progress=78.3% loss=0.049652 lr=1.227658e-06 elapsed=29:05:46 avg_step=1.34s recent_step=1.15s eta=08:02:24 +2026-08-26 09:01:30 | INFO | train rank=0 | step=78360/100000 progress=78.4% loss=0.042068 lr=1.226573e-06 elapsed=29:06:00 avg_step=1.34s recent_step=1.36s eta=08:02:10 +2026-08-26 09:01:45 | INFO | train rank=0 | step=78370/100000 progress=78.4% loss=0.050308 lr=1.225488e-06 elapsed=29:06:15 avg_step=1.34s recent_step=1.53s eta=08:01:57 +2026-08-26 09:01:57 | INFO | train rank=0 | step=78380/100000 progress=78.4% loss=0.053787 lr=1.224404e-06 elapsed=29:06:27 avg_step=1.34s recent_step=1.18s eta=08:01:44 +2026-08-26 09:02:15 | INFO | train rank=0 | step=78390/100000 progress=78.4% loss=0.072601 lr=1.223320e-06 elapsed=29:06:45 avg_step=1.34s recent_step=1.79s eta=08:01:31 +2026-08-26 09:02:30 | INFO | train rank=0 | step=78400/100000 progress=78.4% loss=0.049042 lr=1.222237e-06 elapsed=29:06:59 avg_step=1.34s recent_step=1.49s eta=08:01:18 +2026-08-26 09:02:41 | INFO | train rank=0 | step=78410/100000 progress=78.4% loss=0.033531 lr=1.221154e-06 elapsed=29:07:11 avg_step=1.34s recent_step=1.19s eta=08:01:05 +2026-08-26 09:02:56 | INFO | train rank=0 | step=78420/100000 progress=78.4% loss=0.050629 lr=1.220071e-06 elapsed=29:07:26 avg_step=1.34s recent_step=1.47s eta=08:00:52 +2026-08-26 09:03:12 | INFO | train rank=0 | step=78430/100000 progress=78.4% loss=0.029884 lr=1.218989e-06 elapsed=29:07:42 avg_step=1.34s recent_step=1.60s eta=08:00:39 +2026-08-26 09:03:23 | INFO | train rank=0 | step=78440/100000 progress=78.4% loss=0.066650 lr=1.217907e-06 elapsed=29:07:53 avg_step=1.34s recent_step=1.13s eta=08:00:25 +2026-08-26 09:03:36 | INFO | train rank=0 | step=78450/100000 progress=78.5% loss=0.057098 lr=1.216826e-06 elapsed=29:08:06 avg_step=1.34s recent_step=1.22s eta=08:00:11 +2026-08-26 09:03:47 | INFO | train rank=0 | step=78460/100000 progress=78.5% loss=0.055054 lr=1.215745e-06 elapsed=29:08:17 avg_step=1.34s recent_step=1.15s eta=07:59:58 +2026-08-26 09:04:03 | INFO | train rank=0 | step=78470/100000 progress=78.5% loss=0.033447 lr=1.214665e-06 elapsed=29:08:33 avg_step=1.34s recent_step=1.59s eta=07:59:45 +2026-08-26 09:04:14 | INFO | train rank=0 | step=78480/100000 progress=78.5% loss=0.049683 lr=1.213585e-06 elapsed=29:08:44 avg_step=1.34s recent_step=1.14s eta=07:59:31 +2026-08-26 09:04:30 | INFO | train rank=0 | step=78490/100000 progress=78.5% loss=0.075958 lr=1.212505e-06 elapsed=29:08:59 avg_step=1.34s recent_step=1.51s eta=07:59:18 +2026-08-26 09:04:41 | INFO | train rank=0 | step=78500/100000 progress=78.5% loss=0.034821 lr=1.211426e-06 elapsed=29:09:11 avg_step=1.34s recent_step=1.16s eta=07:59:04 +2026-08-26 09:04:58 | INFO | train rank=0 | step=78510/100000 progress=78.5% loss=0.030388 lr=1.210347e-06 elapsed=29:09:28 avg_step=1.34s recent_step=1.65s eta=07:58:52 +2026-08-26 09:05:10 | INFO | train rank=0 | step=78520/100000 progress=78.5% loss=0.056274 lr=1.209268e-06 elapsed=29:09:40 avg_step=1.34s recent_step=1.21s eta=07:58:38 +2026-08-26 09:05:21 | INFO | train rank=0 | step=78530/100000 progress=78.5% loss=0.091797 lr=1.208190e-06 elapsed=29:09:51 avg_step=1.34s recent_step=1.18s eta=07:58:24 +2026-08-26 09:05:33 | INFO | train rank=0 | step=78540/100000 progress=78.5% loss=0.052399 lr=1.207113e-06 elapsed=29:10:03 avg_step=1.34s recent_step=1.14s eta=07:58:10 +2026-08-26 09:05:45 | INFO | train rank=0 | step=78550/100000 progress=78.5% loss=0.057907 lr=1.206036e-06 elapsed=29:10:15 avg_step=1.34s recent_step=1.18s eta=07:57:56 +2026-08-26 09:05:56 | INFO | train rank=0 | step=78560/100000 progress=78.6% loss=0.044891 lr=1.204959e-06 elapsed=29:10:26 avg_step=1.34s recent_step=1.12s eta=07:57:42 +2026-08-26 09:06:10 | INFO | train rank=0 | step=78570/100000 progress=78.6% loss=0.043686 lr=1.203883e-06 elapsed=29:10:40 avg_step=1.34s recent_step=1.44s eta=07:57:29 +2026-08-26 09:06:26 | INFO | train rank=0 | step=78580/100000 progress=78.6% loss=0.064026 lr=1.202807e-06 elapsed=29:10:56 avg_step=1.34s recent_step=1.61s eta=07:57:17 +2026-08-26 09:06:41 | INFO | train rank=0 | step=78590/100000 progress=78.6% loss=0.046318 lr=1.201731e-06 elapsed=29:11:11 avg_step=1.34s recent_step=1.48s eta=07:57:04 +2026-08-26 09:06:56 | INFO | train rank=0 | step=78600/100000 progress=78.6% loss=0.043610 lr=1.200656e-06 elapsed=29:11:26 avg_step=1.34s recent_step=1.48s eta=07:56:51 +2026-08-26 09:07:08 | INFO | train rank=0 | step=78610/100000 progress=78.6% loss=0.046204 lr=1.199581e-06 elapsed=29:11:37 avg_step=1.34s recent_step=1.16s eta=07:56:37 +2026-08-26 09:07:19 | INFO | train rank=0 | step=78620/100000 progress=78.6% loss=0.043686 lr=1.198507e-06 elapsed=29:11:49 avg_step=1.34s recent_step=1.14s eta=07:56:23 +2026-08-26 09:07:31 | INFO | train rank=0 | step=78630/100000 progress=78.6% loss=0.046341 lr=1.197433e-06 elapsed=29:12:01 avg_step=1.34s recent_step=1.19s eta=07:56:09 +2026-08-26 09:07:44 | INFO | train rank=0 | step=78640/100000 progress=78.6% loss=0.034912 lr=1.196360e-06 elapsed=29:12:13 avg_step=1.34s recent_step=1.27s eta=07:55:56 +2026-08-26 09:07:58 | INFO | train rank=0 | step=78650/100000 progress=78.6% loss=0.052910 lr=1.195287e-06 elapsed=29:12:28 avg_step=1.34s recent_step=1.48s eta=07:55:43 +2026-08-26 09:08:13 | INFO | train rank=0 | step=78660/100000 progress=78.7% loss=0.036537 lr=1.194214e-06 elapsed=29:12:43 avg_step=1.34s recent_step=1.42s eta=07:55:30 +2026-08-26 09:08:24 | INFO | train rank=0 | step=78670/100000 progress=78.7% loss=0.032814 lr=1.193142e-06 elapsed=29:12:54 avg_step=1.34s recent_step=1.19s eta=07:55:16 +2026-08-26 09:08:41 | INFO | train rank=0 | step=78680/100000 progress=78.7% loss=0.028046 lr=1.192070e-06 elapsed=29:13:11 avg_step=1.34s recent_step=1.62s eta=07:55:03 +2026-08-26 09:08:54 | INFO | train rank=0 | step=78690/100000 progress=78.7% loss=0.041946 lr=1.190999e-06 elapsed=29:13:24 avg_step=1.34s recent_step=1.34s eta=07:54:50 +2026-08-26 09:09:06 | INFO | train rank=0 | step=78700/100000 progress=78.7% loss=0.067581 lr=1.189928e-06 elapsed=29:13:36 avg_step=1.34s recent_step=1.16s eta=07:54:36 +2026-08-26 09:09:18 | INFO | train rank=0 | step=78710/100000 progress=78.7% loss=0.034393 lr=1.188858e-06 elapsed=29:13:48 avg_step=1.34s recent_step=1.25s eta=07:54:22 +2026-08-26 09:09:35 | INFO | train rank=0 | step=78720/100000 progress=78.7% loss=0.101379 lr=1.187787e-06 elapsed=29:14:05 avg_step=1.34s recent_step=1.65s eta=07:54:10 +2026-08-26 09:09:46 | INFO | train rank=0 | step=78730/100000 progress=78.7% loss=0.056335 lr=1.186718e-06 elapsed=29:14:16 avg_step=1.34s recent_step=1.15s eta=07:53:56 +2026-08-26 09:09:58 | INFO | train rank=0 | step=78740/100000 progress=78.7% loss=0.037624 lr=1.185649e-06 elapsed=29:14:28 avg_step=1.34s recent_step=1.20s eta=07:53:42 +2026-08-26 09:10:12 | INFO | train rank=0 | step=78750/100000 progress=78.8% loss=0.028679 lr=1.184580e-06 elapsed=29:14:42 avg_step=1.34s recent_step=1.43s eta=07:53:29 +2026-08-26 09:10:27 | INFO | train rank=0 | step=78760/100000 progress=78.8% loss=0.051025 lr=1.183511e-06 elapsed=29:14:57 avg_step=1.34s recent_step=1.44s eta=07:53:16 +2026-08-26 09:10:40 | INFO | train rank=0 | step=78770/100000 progress=78.8% loss=0.053238 lr=1.182443e-06 elapsed=29:15:10 avg_step=1.34s recent_step=1.31s eta=07:53:03 +2026-08-26 09:10:53 | INFO | train rank=0 | step=78780/100000 progress=78.8% loss=0.043823 lr=1.181376e-06 elapsed=29:15:23 avg_step=1.34s recent_step=1.35s eta=07:52:49 +2026-08-26 09:11:07 | INFO | train rank=0 | step=78790/100000 progress=78.8% loss=0.061325 lr=1.180308e-06 elapsed=29:15:37 avg_step=1.34s recent_step=1.41s eta=07:52:36 +2026-08-26 09:11:23 | INFO | train rank=0 | step=78800/100000 progress=78.8% loss=0.039841 lr=1.179242e-06 elapsed=29:15:52 avg_step=1.34s recent_step=1.50s eta=07:52:23 +2026-08-26 09:11:36 | INFO | train rank=0 | step=78810/100000 progress=78.8% loss=0.033783 lr=1.178175e-06 elapsed=29:16:06 avg_step=1.34s recent_step=1.32s eta=07:52:10 +2026-08-26 09:11:49 | INFO | train rank=0 | step=78820/100000 progress=78.8% loss=0.037064 lr=1.177109e-06 elapsed=29:16:19 avg_step=1.34s recent_step=1.36s eta=07:51:56 +2026-08-26 09:12:01 | INFO | train rank=0 | step=78830/100000 progress=78.8% loss=0.052307 lr=1.176044e-06 elapsed=29:16:31 avg_step=1.34s recent_step=1.17s eta=07:51:43 +2026-08-26 09:12:14 | INFO | train rank=0 | step=78840/100000 progress=78.8% loss=0.071198 lr=1.174979e-06 elapsed=29:16:44 avg_step=1.34s recent_step=1.31s eta=07:51:29 +2026-08-26 09:12:31 | INFO | train rank=0 | step=78850/100000 progress=78.8% loss=0.068848 lr=1.173914e-06 elapsed=29:17:01 avg_step=1.34s recent_step=1.67s eta=07:51:17 +2026-08-26 09:12:42 | INFO | train rank=0 | step=78860/100000 progress=78.9% loss=0.036591 lr=1.172850e-06 elapsed=29:17:12 avg_step=1.34s recent_step=1.17s eta=07:51:03 +2026-08-26 09:12:57 | INFO | train rank=0 | step=78870/100000 progress=78.9% loss=0.027985 lr=1.171786e-06 elapsed=29:17:27 avg_step=1.34s recent_step=1.47s eta=07:50:50 +2026-08-26 09:13:13 | INFO | train rank=0 | step=78880/100000 progress=78.9% loss=0.050705 lr=1.170723e-06 elapsed=29:17:42 avg_step=1.34s recent_step=1.54s eta=07:50:37 +2026-08-26 09:13:25 | INFO | train rank=0 | step=78890/100000 progress=78.9% loss=0.044456 lr=1.169660e-06 elapsed=29:17:55 avg_step=1.34s recent_step=1.25s eta=07:50:23 +2026-08-26 09:13:37 | INFO | train rank=0 | step=78900/100000 progress=78.9% loss=0.038658 lr=1.168597e-06 elapsed=29:18:07 avg_step=1.34s recent_step=1.19s eta=07:50:10 +2026-08-26 09:13:50 | INFO | train rank=0 | step=78910/100000 progress=78.9% loss=0.045914 lr=1.167535e-06 elapsed=29:18:20 avg_step=1.34s recent_step=1.28s eta=07:49:56 +2026-08-26 09:14:03 | INFO | train rank=0 | step=78920/100000 progress=78.9% loss=0.036819 lr=1.166473e-06 elapsed=29:18:33 avg_step=1.34s recent_step=1.31s eta=07:49:43 +2026-08-26 09:14:15 | INFO | train rank=0 | step=78930/100000 progress=78.9% loss=0.042877 lr=1.165412e-06 elapsed=29:18:44 avg_step=1.34s recent_step=1.17s eta=07:49:29 +2026-08-26 09:14:26 | INFO | train rank=0 | step=78940/100000 progress=78.9% loss=0.050632 lr=1.164351e-06 elapsed=29:18:56 avg_step=1.34s recent_step=1.18s eta=07:49:15 +2026-08-26 09:14:39 | INFO | train rank=0 | step=78950/100000 progress=79.0% loss=0.044571 lr=1.163291e-06 elapsed=29:19:09 avg_step=1.34s recent_step=1.28s eta=07:49:02 +2026-08-26 09:14:56 | INFO | train rank=0 | step=78960/100000 progress=79.0% loss=0.042351 lr=1.162230e-06 elapsed=29:19:26 avg_step=1.34s recent_step=1.68s eta=07:48:49 +2026-08-26 09:15:10 | INFO | train rank=0 | step=78970/100000 progress=79.0% loss=0.035591 lr=1.161171e-06 elapsed=29:19:40 avg_step=1.34s recent_step=1.45s eta=07:48:36 +2026-08-26 09:15:26 | INFO | train rank=0 | step=78980/100000 progress=79.0% loss=0.048035 lr=1.160112e-06 elapsed=29:19:56 avg_step=1.34s recent_step=1.55s eta=07:48:23 +2026-08-26 09:15:37 | INFO | train rank=0 | step=78990/100000 progress=79.0% loss=0.050797 lr=1.159053e-06 elapsed=29:20:07 avg_step=1.34s recent_step=1.16s eta=07:48:09 +2026-08-26 09:15:53 | INFO | train rank=0 | step=79000/100000 progress=79.0% loss=0.026581 lr=1.157994e-06 elapsed=29:20:23 avg_step=1.34s recent_step=1.59s eta=07:47:57 +2026-08-26 09:16:08 | INFO | train rank=0 | step=79010/100000 progress=79.0% loss=0.064148 lr=1.156936e-06 elapsed=29:20:38 avg_step=1.34s recent_step=1.43s eta=07:47:44 +2026-08-26 09:16:20 | INFO | train rank=0 | step=79020/100000 progress=79.0% loss=0.030384 lr=1.155879e-06 elapsed=29:20:49 avg_step=1.34s recent_step=1.19s eta=07:47:30 +2026-08-26 09:16:35 | INFO | train rank=0 | step=79030/100000 progress=79.0% loss=0.030434 lr=1.154822e-06 elapsed=29:21:05 avg_step=1.34s recent_step=1.51s eta=07:47:17 +2026-08-26 09:16:46 | INFO | train rank=0 | step=79040/100000 progress=79.0% loss=0.038979 lr=1.153765e-06 elapsed=29:21:16 avg_step=1.34s recent_step=1.12s eta=07:47:03 +2026-08-26 09:16:58 | INFO | train rank=0 | step=79050/100000 progress=79.0% loss=0.040604 lr=1.152709e-06 elapsed=29:21:28 avg_step=1.34s recent_step=1.25s eta=07:46:49 +2026-08-26 09:17:11 | INFO | train rank=0 | step=79060/100000 progress=79.1% loss=0.037117 lr=1.151653e-06 elapsed=29:21:41 avg_step=1.34s recent_step=1.22s eta=07:46:36 +2026-08-26 09:17:25 | INFO | train rank=0 | step=79070/100000 progress=79.1% loss=0.041458 lr=1.150598e-06 elapsed=29:21:55 avg_step=1.34s recent_step=1.46s eta=07:46:23 +2026-08-26 09:17:37 | INFO | train rank=0 | step=79080/100000 progress=79.1% loss=0.076599 lr=1.149543e-06 elapsed=29:22:07 avg_step=1.34s recent_step=1.19s eta=07:46:09 +2026-08-26 09:17:50 | INFO | train rank=0 | step=79090/100000 progress=79.1% loss=0.023201 lr=1.148488e-06 elapsed=29:22:20 avg_step=1.34s recent_step=1.32s eta=07:45:55 +2026-08-26 09:18:05 | INFO | train rank=0 | step=79100/100000 progress=79.1% loss=0.040298 lr=1.147434e-06 elapsed=29:22:35 avg_step=1.34s recent_step=1.50s eta=07:45:43 +2026-08-26 09:18:17 | INFO | train rank=0 | step=79110/100000 progress=79.1% loss=0.054321 lr=1.146380e-06 elapsed=29:22:47 avg_step=1.34s recent_step=1.14s eta=07:45:29 +2026-08-26 09:18:29 | INFO | train rank=0 | step=79120/100000 progress=79.1% loss=0.033813 lr=1.145327e-06 elapsed=29:22:59 avg_step=1.34s recent_step=1.25s eta=07:45:15 +2026-08-26 09:18:45 | INFO | train rank=0 | step=79130/100000 progress=79.1% loss=0.048508 lr=1.144274e-06 elapsed=29:23:15 avg_step=1.34s recent_step=1.56s eta=07:45:02 +2026-08-26 09:18:57 | INFO | train rank=0 | step=79140/100000 progress=79.1% loss=0.038666 lr=1.143221e-06 elapsed=29:23:27 avg_step=1.34s recent_step=1.21s eta=07:44:49 +2026-08-26 09:19:12 | INFO | train rank=0 | step=79150/100000 progress=79.1% loss=0.050995 lr=1.142169e-06 elapsed=29:23:42 avg_step=1.34s recent_step=1.50s eta=07:44:36 +2026-08-26 09:19:25 | INFO | train rank=0 | step=79160/100000 progress=79.2% loss=0.047470 lr=1.141118e-06 elapsed=29:23:55 avg_step=1.34s recent_step=1.35s eta=07:44:22 +2026-08-26 09:19:37 | INFO | train rank=0 | step=79170/100000 progress=79.2% loss=0.036827 lr=1.140066e-06 elapsed=29:24:07 avg_step=1.34s recent_step=1.13s eta=07:44:08 +2026-08-26 09:19:49 | INFO | train rank=0 | step=79180/100000 progress=79.2% loss=0.050949 lr=1.139016e-06 elapsed=29:24:18 avg_step=1.34s recent_step=1.19s eta=07:43:55 +2026-08-26 09:20:06 | INFO | train rank=0 | step=79190/100000 progress=79.2% loss=0.061646 lr=1.137965e-06 elapsed=29:24:36 avg_step=1.34s recent_step=1.76s eta=07:43:42 +2026-08-26 09:20:18 | INFO | train rank=0 | step=79200/100000 progress=79.2% loss=0.079895 lr=1.136915e-06 elapsed=29:24:48 avg_step=1.34s recent_step=1.17s eta=07:43:29 +2026-08-26 09:20:37 | INFO | train rank=0 | step=79210/100000 progress=79.2% loss=0.032539 lr=1.135866e-06 elapsed=29:25:07 avg_step=1.34s recent_step=1.89s eta=07:43:17 +2026-08-26 09:20:48 | INFO | train rank=0 | step=79220/100000 progress=79.2% loss=0.039932 lr=1.134817e-06 elapsed=29:25:18 avg_step=1.34s recent_step=1.10s eta=07:43:03 +2026-08-26 09:21:02 | INFO | train rank=0 | step=79230/100000 progress=79.2% loss=0.035980 lr=1.133768e-06 elapsed=29:25:32 avg_step=1.34s recent_step=1.47s eta=07:42:50 +2026-08-26 09:21:17 | INFO | train rank=0 | step=79240/100000 progress=79.2% loss=0.034988 lr=1.132720e-06 elapsed=29:25:47 avg_step=1.34s recent_step=1.43s eta=07:42:36 +2026-08-26 09:21:28 | INFO | train rank=0 | step=79250/100000 progress=79.2% loss=0.068695 lr=1.131672e-06 elapsed=29:25:58 avg_step=1.34s recent_step=1.15s eta=07:42:23 +2026-08-26 09:21:40 | INFO | train rank=0 | step=79260/100000 progress=79.3% loss=0.056015 lr=1.130624e-06 elapsed=29:26:10 avg_step=1.34s recent_step=1.18s eta=07:42:09 +2026-08-26 09:21:51 | INFO | train rank=0 | step=79270/100000 progress=79.3% loss=0.031860 lr=1.129577e-06 elapsed=29:26:21 avg_step=1.34s recent_step=1.08s eta=07:41:55 +2026-08-26 09:22:03 | INFO | train rank=0 | step=79280/100000 progress=79.3% loss=0.045135 lr=1.128531e-06 elapsed=29:26:33 avg_step=1.34s recent_step=1.25s eta=07:41:41 +2026-08-26 09:22:20 | INFO | train rank=0 | step=79290/100000 progress=79.3% loss=0.042694 lr=1.127485e-06 elapsed=29:26:49 avg_step=1.34s recent_step=1.62s eta=07:41:29 +2026-08-26 09:22:31 | INFO | train rank=0 | step=79300/100000 progress=79.3% loss=0.028931 lr=1.126439e-06 elapsed=29:27:01 avg_step=1.34s recent_step=1.17s eta=07:41:15 +2026-08-26 09:22:45 | INFO | train rank=0 | step=79310/100000 progress=79.3% loss=0.033432 lr=1.125394e-06 elapsed=29:27:14 avg_step=1.34s recent_step=1.33s eta=07:41:01 +2026-08-26 09:22:56 | INFO | train rank=0 | step=79320/100000 progress=79.3% loss=0.061768 lr=1.124349e-06 elapsed=29:27:26 avg_step=1.34s recent_step=1.11s eta=07:40:47 +2026-08-26 09:23:13 | INFO | train rank=0 | step=79330/100000 progress=79.3% loss=0.074036 lr=1.123304e-06 elapsed=29:27:43 avg_step=1.34s recent_step=1.70s eta=07:40:35 +2026-08-26 09:23:25 | INFO | train rank=0 | step=79340/100000 progress=79.3% loss=0.029724 lr=1.122260e-06 elapsed=29:27:55 avg_step=1.34s recent_step=1.27s eta=07:40:21 +2026-08-26 09:23:43 | INFO | train rank=0 | step=79350/100000 progress=79.3% loss=0.031448 lr=1.121217e-06 elapsed=29:28:13 avg_step=1.34s recent_step=1.79s eta=07:40:09 +2026-08-26 09:23:55 | INFO | train rank=0 | step=79360/100000 progress=79.4% loss=0.043549 lr=1.120174e-06 elapsed=29:28:25 avg_step=1.34s recent_step=1.16s eta=07:39:55 +2026-08-26 09:24:06 | INFO | train rank=0 | step=79370/100000 progress=79.4% loss=0.061615 lr=1.119131e-06 elapsed=29:28:36 avg_step=1.34s recent_step=1.13s eta=07:39:42 +2026-08-26 09:24:23 | INFO | train rank=0 | step=79380/100000 progress=79.4% loss=0.064026 lr=1.118088e-06 elapsed=29:28:53 avg_step=1.34s recent_step=1.69s eta=07:39:29 +2026-08-26 09:24:35 | INFO | train rank=0 | step=79390/100000 progress=79.4% loss=0.050644 lr=1.117047e-06 elapsed=29:29:05 avg_step=1.34s recent_step=1.17s eta=07:39:15 +2026-08-26 09:24:46 | INFO | train rank=0 | step=79400/100000 progress=79.4% loss=0.040932 lr=1.116005e-06 elapsed=29:29:16 avg_step=1.34s recent_step=1.15s eta=07:39:01 +2026-08-26 09:25:00 | INFO | train rank=0 | step=79410/100000 progress=79.4% loss=0.024834 lr=1.114964e-06 elapsed=29:29:30 avg_step=1.34s recent_step=1.38s eta=07:38:48 +2026-08-26 09:25:13 | INFO | train rank=0 | step=79420/100000 progress=79.4% loss=0.031128 lr=1.113923e-06 elapsed=29:29:43 avg_step=1.34s recent_step=1.26s eta=07:38:35 +2026-08-26 09:25:29 | INFO | train rank=0 | step=79430/100000 progress=79.4% loss=0.035950 lr=1.112883e-06 elapsed=29:29:59 avg_step=1.34s recent_step=1.63s eta=07:38:22 +2026-08-26 09:25:41 | INFO | train rank=0 | step=79440/100000 progress=79.4% loss=0.039841 lr=1.111843e-06 elapsed=29:30:11 avg_step=1.34s recent_step=1.18s eta=07:38:08 +2026-08-26 09:25:52 | INFO | train rank=0 | step=79450/100000 progress=79.5% loss=0.061874 lr=1.110804e-06 elapsed=29:30:22 avg_step=1.34s recent_step=1.16s eta=07:37:54 +2026-08-26 09:26:05 | INFO | train rank=0 | step=79460/100000 progress=79.5% loss=0.029976 lr=1.109765e-06 elapsed=29:30:35 avg_step=1.34s recent_step=1.22s eta=07:37:41 +2026-08-26 09:26:19 | INFO | train rank=0 | step=79470/100000 progress=79.5% loss=0.041351 lr=1.108727e-06 elapsed=29:30:49 avg_step=1.34s recent_step=1.47s eta=07:37:28 +2026-08-26 09:26:33 | INFO | train rank=0 | step=79480/100000 progress=79.5% loss=0.031136 lr=1.107688e-06 elapsed=29:31:03 avg_step=1.34s recent_step=1.41s eta=07:37:15 +2026-08-26 09:26:46 | INFO | train rank=0 | step=79490/100000 progress=79.5% loss=0.024086 lr=1.106651e-06 elapsed=29:31:16 avg_step=1.34s recent_step=1.22s eta=07:37:01 +2026-08-26 09:27:01 | INFO | train rank=0 | step=79500/100000 progress=79.5% loss=0.049316 lr=1.105614e-06 elapsed=29:31:31 avg_step=1.34s recent_step=1.51s eta=07:36:48 +2026-08-26 09:27:15 | INFO | train rank=0 | step=79510/100000 progress=79.5% loss=0.068146 lr=1.104577e-06 elapsed=29:31:45 avg_step=1.34s recent_step=1.46s eta=07:36:35 +2026-08-26 09:27:31 | INFO | train rank=0 | step=79520/100000 progress=79.5% loss=0.079224 lr=1.103540e-06 elapsed=29:32:01 avg_step=1.34s recent_step=1.56s eta=07:36:22 +2026-08-26 09:27:46 | INFO | train rank=0 | step=79530/100000 progress=79.5% loss=0.044975 lr=1.102504e-06 elapsed=29:32:16 avg_step=1.34s recent_step=1.48s eta=07:36:09 +2026-08-26 09:27:57 | INFO | train rank=0 | step=79540/100000 progress=79.5% loss=0.033112 lr=1.101469e-06 elapsed=29:32:27 avg_step=1.34s recent_step=1.15s eta=07:35:55 +2026-08-26 09:28:13 | INFO | train rank=0 | step=79550/100000 progress=79.5% loss=0.029854 lr=1.100434e-06 elapsed=29:32:43 avg_step=1.34s recent_step=1.54s eta=07:35:42 +2026-08-26 09:28:24 | INFO | train rank=0 | step=79560/100000 progress=79.6% loss=0.055298 lr=1.099399e-06 elapsed=29:32:54 avg_step=1.34s recent_step=1.16s eta=07:35:29 +2026-08-26 09:28:39 | INFO | train rank=0 | step=79570/100000 progress=79.6% loss=0.084167 lr=1.098365e-06 elapsed=29:33:09 avg_step=1.34s recent_step=1.46s eta=07:35:15 +2026-08-26 09:28:51 | INFO | train rank=0 | step=79580/100000 progress=79.6% loss=0.037659 lr=1.097331e-06 elapsed=29:33:20 avg_step=1.34s recent_step=1.17s eta=07:35:02 +2026-08-26 09:29:04 | INFO | train rank=0 | step=79590/100000 progress=79.6% loss=0.031662 lr=1.096298e-06 elapsed=29:33:34 avg_step=1.34s recent_step=1.38s eta=07:34:48 +2026-08-26 09:29:16 | INFO | train rank=0 | step=79600/100000 progress=79.6% loss=0.040649 lr=1.095265e-06 elapsed=29:33:46 avg_step=1.34s recent_step=1.20s eta=07:34:35 +2026-08-26 09:29:28 | INFO | train rank=0 | step=79610/100000 progress=79.6% loss=0.061966 lr=1.094232e-06 elapsed=29:33:57 avg_step=1.34s recent_step=1.12s eta=07:34:21 +2026-08-26 09:29:45 | INFO | train rank=0 | step=79620/100000 progress=79.6% loss=0.050568 lr=1.093200e-06 elapsed=29:34:15 avg_step=1.34s recent_step=1.74s eta=07:34:08 +2026-08-26 09:30:00 | INFO | train rank=0 | step=79630/100000 progress=79.6% loss=0.034500 lr=1.092168e-06 elapsed=29:34:30 avg_step=1.34s recent_step=1.48s eta=07:33:55 +2026-08-26 09:30:15 | INFO | train rank=0 | step=79640/100000 progress=79.6% loss=0.080200 lr=1.091137e-06 elapsed=29:34:45 avg_step=1.34s recent_step=1.50s eta=07:33:42 +2026-08-26 09:30:27 | INFO | train rank=0 | step=79650/100000 progress=79.7% loss=0.061462 lr=1.090106e-06 elapsed=29:34:56 avg_step=1.34s recent_step=1.18s eta=07:33:29 +2026-08-26 09:30:41 | INFO | train rank=0 | step=79660/100000 progress=79.7% loss=0.044495 lr=1.089076e-06 elapsed=29:35:11 avg_step=1.34s recent_step=1.42s eta=07:33:16 +2026-08-26 09:30:54 | INFO | train rank=0 | step=79670/100000 progress=79.7% loss=0.036972 lr=1.088046e-06 elapsed=29:35:24 avg_step=1.34s recent_step=1.32s eta=07:33:02 +2026-08-26 09:31:09 | INFO | train rank=0 | step=79680/100000 progress=79.7% loss=0.045792 lr=1.087016e-06 elapsed=29:35:39 avg_step=1.34s recent_step=1.48s eta=07:32:49 +2026-08-26 09:31:23 | INFO | train rank=0 | step=79690/100000 progress=79.7% loss=0.039093 lr=1.085987e-06 elapsed=29:35:53 avg_step=1.34s recent_step=1.39s eta=07:32:36 +2026-08-26 09:31:37 | INFO | train rank=0 | step=79700/100000 progress=79.7% loss=0.036331 lr=1.084959e-06 elapsed=29:36:06 avg_step=1.34s recent_step=1.39s eta=07:32:23 +2026-08-26 09:31:54 | INFO | train rank=0 | step=79710/100000 progress=79.7% loss=0.070648 lr=1.083930e-06 elapsed=29:36:24 avg_step=1.34s recent_step=1.78s eta=07:32:10 +2026-08-26 09:32:08 | INFO | train rank=0 | step=79720/100000 progress=79.7% loss=0.045799 lr=1.082902e-06 elapsed=29:36:38 avg_step=1.34s recent_step=1.39s eta=07:31:57 +2026-08-26 09:32:20 | INFO | train rank=0 | step=79730/100000 progress=79.7% loss=0.035217 lr=1.081875e-06 elapsed=29:36:50 avg_step=1.34s recent_step=1.14s eta=07:31:43 +2026-08-26 09:32:34 | INFO | train rank=0 | step=79740/100000 progress=79.7% loss=0.034874 lr=1.080848e-06 elapsed=29:37:04 avg_step=1.34s recent_step=1.40s eta=07:31:30 +2026-08-26 09:32:50 | INFO | train rank=0 | step=79750/100000 progress=79.8% loss=0.069275 lr=1.079822e-06 elapsed=29:37:19 avg_step=1.34s recent_step=1.59s eta=07:31:17 +2026-08-26 09:33:01 | INFO | train rank=0 | step=79760/100000 progress=79.8% loss=0.028191 lr=1.078795e-06 elapsed=29:37:30 avg_step=1.34s recent_step=1.10s eta=07:31:03 +2026-08-26 09:33:12 | INFO | train rank=0 | step=79770/100000 progress=79.8% loss=0.051224 lr=1.077770e-06 elapsed=29:37:42 avg_step=1.34s recent_step=1.11s eta=07:30:49 +2026-08-26 09:33:26 | INFO | train rank=0 | step=79780/100000 progress=79.8% loss=0.037445 lr=1.076744e-06 elapsed=29:37:56 avg_step=1.34s recent_step=1.44s eta=07:30:36 +2026-08-26 09:33:38 | INFO | train rank=0 | step=79790/100000 progress=79.8% loss=0.074341 lr=1.075720e-06 elapsed=29:38:08 avg_step=1.34s recent_step=1.18s eta=07:30:23 +2026-08-26 09:33:52 | INFO | train rank=0 | step=79800/100000 progress=79.8% loss=0.061768 lr=1.074695e-06 elapsed=29:38:21 avg_step=1.34s recent_step=1.37s eta=07:30:09 +2026-08-26 09:34:03 | INFO | train rank=0 | step=79810/100000 progress=79.8% loss=0.042374 lr=1.073671e-06 elapsed=29:38:33 avg_step=1.34s recent_step=1.17s eta=07:29:55 +2026-08-26 09:34:18 | INFO | train rank=0 | step=79820/100000 progress=79.8% loss=0.052704 lr=1.072648e-06 elapsed=29:38:48 avg_step=1.34s recent_step=1.46s eta=07:29:42 +2026-08-26 09:34:30 | INFO | train rank=0 | step=79830/100000 progress=79.8% loss=0.044823 lr=1.071625e-06 elapsed=29:39:00 avg_step=1.34s recent_step=1.22s eta=07:29:29 +2026-08-26 09:34:42 | INFO | train rank=0 | step=79840/100000 progress=79.8% loss=0.074554 lr=1.070602e-06 elapsed=29:39:12 avg_step=1.34s recent_step=1.16s eta=07:29:15 +2026-08-26 09:34:55 | INFO | train rank=0 | step=79850/100000 progress=79.8% loss=0.037262 lr=1.069580e-06 elapsed=29:39:25 avg_step=1.34s recent_step=1.35s eta=07:29:02 +2026-08-26 09:35:10 | INFO | train rank=0 | step=79860/100000 progress=79.9% loss=0.051956 lr=1.068558e-06 elapsed=29:39:40 avg_step=1.34s recent_step=1.52s eta=07:28:49 +2026-08-26 09:35:24 | INFO | train rank=0 | step=79870/100000 progress=79.9% loss=0.053497 lr=1.067536e-06 elapsed=29:39:54 avg_step=1.34s recent_step=1.41s eta=07:28:35 +2026-08-26 09:35:43 | INFO | train rank=0 | step=79880/100000 progress=79.9% loss=0.038177 lr=1.066515e-06 elapsed=29:40:13 avg_step=1.34s recent_step=1.82s eta=07:28:23 +2026-08-26 09:35:55 | INFO | train rank=0 | step=79890/100000 progress=79.9% loss=0.054642 lr=1.065495e-06 elapsed=29:40:25 avg_step=1.34s recent_step=1.20s eta=07:28:10 +2026-08-26 09:36:07 | INFO | train rank=0 | step=79900/100000 progress=79.9% loss=0.039284 lr=1.064475e-06 elapsed=29:40:37 avg_step=1.34s recent_step=1.24s eta=07:27:56 +2026-08-26 09:36:22 | INFO | train rank=0 | step=79910/100000 progress=79.9% loss=0.112427 lr=1.063455e-06 elapsed=29:40:52 avg_step=1.34s recent_step=1.53s eta=07:27:43 +2026-08-26 09:36:38 | INFO | train rank=0 | step=79920/100000 progress=79.9% loss=0.062408 lr=1.062436e-06 elapsed=29:41:08 avg_step=1.34s recent_step=1.56s eta=07:27:30 +2026-08-26 09:36:50 | INFO | train rank=0 | step=79930/100000 progress=79.9% loss=0.044983 lr=1.061417e-06 elapsed=29:41:19 avg_step=1.34s recent_step=1.15s eta=07:27:16 +2026-08-26 09:37:01 | INFO | train rank=0 | step=79940/100000 progress=79.9% loss=0.053268 lr=1.060399e-06 elapsed=29:41:31 avg_step=1.34s recent_step=1.19s eta=07:27:03 +2026-08-26 09:37:14 | INFO | train rank=0 | step=79950/100000 progress=80.0% loss=0.038902 lr=1.059381e-06 elapsed=29:41:43 avg_step=1.34s recent_step=1.22s eta=07:26:49 +2026-08-26 09:37:25 | INFO | train rank=0 | step=79960/100000 progress=80.0% loss=0.064819 lr=1.058363e-06 elapsed=29:41:55 avg_step=1.34s recent_step=1.16s eta=07:26:35 +2026-08-26 09:37:38 | INFO | train rank=0 | step=79970/100000 progress=80.0% loss=0.039673 lr=1.057346e-06 elapsed=29:42:08 avg_step=1.34s recent_step=1.25s eta=07:26:22 +2026-08-26 09:37:49 | INFO | train rank=0 | step=79980/100000 progress=80.0% loss=0.047867 lr=1.056329e-06 elapsed=29:42:19 avg_step=1.34s recent_step=1.14s eta=07:26:08 +2026-08-26 09:38:05 | INFO | train rank=0 | step=79990/100000 progress=80.0% loss=0.051041 lr=1.055313e-06 elapsed=29:42:35 avg_step=1.34s recent_step=1.55s eta=07:25:55 +2026-08-26 09:38:17 | INFO | train rank=0 | step=80000/100000 progress=80.0% loss=0.020306 lr=1.054297e-06 elapsed=29:42:47 avg_step=1.34s recent_step=1.21s eta=07:25:41 +2026-08-26 09:38:31 | INFO | train rank=0 | step=80010/100000 progress=80.0% loss=0.042679 lr=1.053282e-06 elapsed=29:43:01 avg_step=1.34s recent_step=1.43s eta=07:25:28 +2026-08-26 09:38:44 | INFO | train rank=0 | step=80020/100000 progress=80.0% loss=0.049011 lr=1.052267e-06 elapsed=29:43:14 avg_step=1.34s recent_step=1.27s eta=07:25:15 +2026-08-26 09:39:00 | INFO | train rank=0 | step=80030/100000 progress=80.0% loss=0.058914 lr=1.051253e-06 elapsed=29:43:30 avg_step=1.34s recent_step=1.66s eta=07:25:02 +2026-08-26 09:39:12 | INFO | train rank=0 | step=80040/100000 progress=80.0% loss=0.051376 lr=1.050239e-06 elapsed=29:43:42 avg_step=1.34s recent_step=1.19s eta=07:24:48 +2026-08-26 09:39:24 | INFO | train rank=0 | step=80050/100000 progress=80.0% loss=0.034447 lr=1.049225e-06 elapsed=29:43:54 avg_step=1.34s recent_step=1.13s eta=07:24:34 +2026-08-26 09:39:35 | INFO | train rank=0 | step=80060/100000 progress=80.1% loss=0.043236 lr=1.048212e-06 elapsed=29:44:05 avg_step=1.34s recent_step=1.16s eta=07:24:21 +2026-08-26 09:39:51 | INFO | train rank=0 | step=80070/100000 progress=80.1% loss=0.047935 lr=1.047199e-06 elapsed=29:44:21 avg_step=1.34s recent_step=1.53s eta=07:24:08 +2026-08-26 09:40:03 | INFO | train rank=0 | step=80080/100000 progress=80.1% loss=0.049042 lr=1.046187e-06 elapsed=29:44:33 avg_step=1.34s recent_step=1.26s eta=07:23:54 +2026-08-26 09:40:20 | INFO | train rank=0 | step=80090/100000 progress=80.1% loss=0.045334 lr=1.045175e-06 elapsed=29:44:50 avg_step=1.34s recent_step=1.67s eta=07:23:42 +2026-08-26 09:40:34 | INFO | train rank=0 | step=80100/100000 progress=80.1% loss=0.061989 lr=1.044163e-06 elapsed=29:45:04 avg_step=1.34s recent_step=1.37s eta=07:23:28 +2026-08-26 09:40:46 | INFO | train rank=0 | step=80110/100000 progress=80.1% loss=0.048737 lr=1.043152e-06 elapsed=29:45:16 avg_step=1.34s recent_step=1.27s eta=07:23:15 +2026-08-26 09:40:59 | INFO | train rank=0 | step=80120/100000 progress=80.1% loss=0.044205 lr=1.042142e-06 elapsed=29:45:29 avg_step=1.34s recent_step=1.29s eta=07:23:01 +2026-08-26 09:41:14 | INFO | train rank=0 | step=80130/100000 progress=80.1% loss=0.121277 lr=1.041131e-06 elapsed=29:45:44 avg_step=1.34s recent_step=1.50s eta=07:22:48 +2026-08-26 09:41:26 | INFO | train rank=0 | step=80140/100000 progress=80.1% loss=0.048019 lr=1.040122e-06 elapsed=29:45:55 avg_step=1.34s recent_step=1.13s eta=07:22:34 +2026-08-26 09:41:39 | INFO | train rank=0 | step=80150/100000 progress=80.2% loss=0.045731 lr=1.039112e-06 elapsed=29:46:09 avg_step=1.34s recent_step=1.38s eta=07:22:21 +2026-08-26 09:41:54 | INFO | train rank=0 | step=80160/100000 progress=80.2% loss=0.063324 lr=1.038103e-06 elapsed=29:46:24 avg_step=1.34s recent_step=1.44s eta=07:22:08 +2026-08-26 09:42:05 | INFO | train rank=0 | step=80170/100000 progress=80.2% loss=0.058655 lr=1.037095e-06 elapsed=29:46:35 avg_step=1.34s recent_step=1.18s eta=07:21:54 +2026-08-26 09:42:17 | INFO | train rank=0 | step=80180/100000 progress=80.2% loss=0.047272 lr=1.036087e-06 elapsed=29:46:47 avg_step=1.34s recent_step=1.16s eta=07:21:41 +2026-08-26 09:42:32 | INFO | train rank=0 | step=80190/100000 progress=80.2% loss=0.065704 lr=1.035079e-06 elapsed=29:47:02 avg_step=1.34s recent_step=1.46s eta=07:21:27 +2026-08-26 09:42:45 | INFO | train rank=0 | step=80200/100000 progress=80.2% loss=0.030121 lr=1.034072e-06 elapsed=29:47:15 avg_step=1.34s recent_step=1.37s eta=07:21:14 +2026-08-26 09:42:58 | INFO | train rank=0 | step=80210/100000 progress=80.2% loss=0.050064 lr=1.033066e-06 elapsed=29:47:28 avg_step=1.34s recent_step=1.25s eta=07:21:01 +2026-08-26 09:43:13 | INFO | train rank=0 | step=80220/100000 progress=80.2% loss=0.041313 lr=1.032059e-06 elapsed=29:47:43 avg_step=1.34s recent_step=1.49s eta=07:20:48 +2026-08-26 09:43:29 | INFO | train rank=0 | step=80230/100000 progress=80.2% loss=0.039337 lr=1.031053e-06 elapsed=29:47:59 avg_step=1.34s recent_step=1.65s eta=07:20:35 +2026-08-26 09:43:44 | INFO | train rank=0 | step=80240/100000 progress=80.2% loss=0.042526 lr=1.030048e-06 elapsed=29:48:14 avg_step=1.34s recent_step=1.44s eta=07:20:22 +2026-08-26 09:43:56 | INFO | train rank=0 | step=80250/100000 progress=80.2% loss=0.070633 lr=1.029043e-06 elapsed=29:48:26 avg_step=1.34s recent_step=1.22s eta=07:20:08 +2026-08-26 09:44:13 | INFO | train rank=0 | step=80260/100000 progress=80.3% loss=0.036255 lr=1.028038e-06 elapsed=29:48:43 avg_step=1.34s recent_step=1.71s eta=07:19:56 +2026-08-26 09:44:25 | INFO | train rank=0 | step=80270/100000 progress=80.3% loss=0.049698 lr=1.027034e-06 elapsed=29:48:55 avg_step=1.34s recent_step=1.23s eta=07:19:42 +2026-08-26 09:44:38 | INFO | train rank=0 | step=80280/100000 progress=80.3% loss=0.061310 lr=1.026031e-06 elapsed=29:49:08 avg_step=1.34s recent_step=1.27s eta=07:19:29 +2026-08-26 09:44:49 | INFO | train rank=0 | step=80290/100000 progress=80.3% loss=0.041611 lr=1.025027e-06 elapsed=29:49:19 avg_step=1.34s recent_step=1.15s eta=07:19:15 +2026-08-26 09:45:03 | INFO | train rank=0 | step=80300/100000 progress=80.3% loss=0.065384 lr=1.024025e-06 elapsed=29:49:33 avg_step=1.34s recent_step=1.32s eta=07:19:01 +2026-08-26 09:45:16 | INFO | train rank=0 | step=80310/100000 progress=80.3% loss=0.030357 lr=1.023022e-06 elapsed=29:49:46 avg_step=1.34s recent_step=1.33s eta=07:18:48 +2026-08-26 09:45:29 | INFO | train rank=0 | step=80320/100000 progress=80.3% loss=0.048965 lr=1.022020e-06 elapsed=29:49:59 avg_step=1.34s recent_step=1.33s eta=07:18:35 +2026-08-26 09:45:46 | INFO | train rank=0 | step=80330/100000 progress=80.3% loss=0.033966 lr=1.021019e-06 elapsed=29:50:16 avg_step=1.34s recent_step=1.64s eta=07:18:22 +2026-08-26 09:45:58 | INFO | train rank=0 | step=80340/100000 progress=80.3% loss=0.095276 lr=1.020018e-06 elapsed=29:50:28 avg_step=1.34s recent_step=1.20s eta=07:18:08 +2026-08-26 09:46:09 | INFO | train rank=0 | step=80350/100000 progress=80.3% loss=0.033829 lr=1.019017e-06 elapsed=29:50:39 avg_step=1.34s recent_step=1.15s eta=07:17:54 +2026-08-26 09:46:22 | INFO | train rank=0 | step=80360/100000 progress=80.4% loss=0.062805 lr=1.018017e-06 elapsed=29:50:52 avg_step=1.34s recent_step=1.25s eta=07:17:41 +2026-08-26 09:46:36 | INFO | train rank=0 | step=80370/100000 progress=80.4% loss=0.036896 lr=1.017017e-06 elapsed=29:51:06 avg_step=1.34s recent_step=1.47s eta=07:17:28 +2026-08-26 09:46:50 | INFO | train rank=0 | step=80380/100000 progress=80.4% loss=0.058380 lr=1.016018e-06 elapsed=29:51:20 avg_step=1.34s recent_step=1.35s eta=07:17:14 +2026-08-26 09:47:03 | INFO | train rank=0 | step=80390/100000 progress=80.4% loss=0.056519 lr=1.015019e-06 elapsed=29:51:32 avg_step=1.34s recent_step=1.26s eta=07:17:01 +2026-08-26 09:47:21 | INFO | train rank=0 | step=80400/100000 progress=80.4% loss=0.035767 lr=1.014021e-06 elapsed=29:51:51 avg_step=1.34s recent_step=1.84s eta=07:16:49 +2026-08-26 09:47:33 | INFO | train rank=0 | step=80410/100000 progress=80.4% loss=0.099915 lr=1.013023e-06 elapsed=29:52:02 avg_step=1.34s recent_step=1.16s eta=07:16:35 +2026-08-26 09:47:45 | INFO | train rank=0 | step=80420/100000 progress=80.4% loss=0.056580 lr=1.012025e-06 elapsed=29:52:15 avg_step=1.34s recent_step=1.25s eta=07:16:21 +2026-08-26 09:47:59 | INFO | train rank=0 | step=80430/100000 progress=80.4% loss=0.043869 lr=1.011028e-06 elapsed=29:52:29 avg_step=1.34s recent_step=1.39s eta=07:16:08 +2026-08-26 09:48:12 | INFO | train rank=0 | step=80440/100000 progress=80.4% loss=0.039780 lr=1.010031e-06 elapsed=29:52:42 avg_step=1.34s recent_step=1.36s eta=07:15:55 +2026-08-26 09:48:26 | INFO | train rank=0 | step=80450/100000 progress=80.5% loss=0.049225 lr=1.009035e-06 elapsed=29:52:56 avg_step=1.34s recent_step=1.36s eta=07:15:41 +2026-08-26 09:48:42 | INFO | train rank=0 | step=80460/100000 progress=80.5% loss=0.055824 lr=1.008039e-06 elapsed=29:53:11 avg_step=1.34s recent_step=1.55s eta=07:15:29 +2026-08-26 09:48:55 | INFO | train rank=0 | step=80470/100000 progress=80.5% loss=0.040680 lr=1.007044e-06 elapsed=29:53:25 avg_step=1.34s recent_step=1.38s eta=07:15:15 +2026-08-26 09:49:08 | INFO | train rank=0 | step=80480/100000 progress=80.5% loss=0.041031 lr=1.006049e-06 elapsed=29:53:38 avg_step=1.34s recent_step=1.26s eta=07:15:02 +2026-08-26 09:49:24 | INFO | train rank=0 | step=80490/100000 progress=80.5% loss=0.038315 lr=1.005054e-06 elapsed=29:53:53 avg_step=1.34s recent_step=1.55s eta=07:14:49 +2026-08-26 09:49:40 | INFO | train rank=0 | step=80500/100000 progress=80.5% loss=0.023472 lr=1.004060e-06 elapsed=29:54:10 avg_step=1.34s recent_step=1.66s eta=07:14:36 +2026-08-26 09:49:54 | INFO | train rank=0 | step=80510/100000 progress=80.5% loss=0.042725 lr=1.003066e-06 elapsed=29:54:24 avg_step=1.34s recent_step=1.36s eta=07:14:23 +2026-08-26 09:50:05 | INFO | train rank=0 | step=80520/100000 progress=80.5% loss=0.036026 lr=1.002073e-06 elapsed=29:54:35 avg_step=1.34s recent_step=1.15s eta=07:14:09 +2026-08-26 09:50:17 | INFO | train rank=0 | step=80530/100000 progress=80.5% loss=0.028233 lr=1.001080e-06 elapsed=29:54:47 avg_step=1.34s recent_step=1.16s eta=07:13:55 +2026-08-26 09:50:29 | INFO | train rank=0 | step=80540/100000 progress=80.5% loss=0.045105 lr=1.000088e-06 elapsed=29:54:59 avg_step=1.34s recent_step=1.21s eta=07:13:42 +2026-08-26 09:50:41 | INFO | train rank=0 | step=80550/100000 progress=80.5% loss=0.064850 lr=9.990961e-07 elapsed=29:55:11 avg_step=1.34s recent_step=1.24s eta=07:13:28 +2026-08-26 09:50:54 | INFO | train rank=0 | step=80560/100000 progress=80.6% loss=0.054062 lr=9.981047e-07 elapsed=29:55:24 avg_step=1.34s recent_step=1.28s eta=07:13:15 +2026-08-26 09:51:06 | INFO | train rank=0 | step=80570/100000 progress=80.6% loss=0.029137 lr=9.971136e-07 elapsed=29:55:36 avg_step=1.34s recent_step=1.15s eta=07:13:01 +2026-08-26 09:51:21 | INFO | train rank=0 | step=80580/100000 progress=80.6% loss=0.077484 lr=9.961231e-07 elapsed=29:55:51 avg_step=1.34s recent_step=1.53s eta=07:12:48 +2026-08-26 09:51:33 | INFO | train rank=0 | step=80590/100000 progress=80.6% loss=0.022621 lr=9.951329e-07 elapsed=29:56:03 avg_step=1.34s recent_step=1.17s eta=07:12:34 +2026-08-26 09:51:44 | INFO | train rank=0 | step=80600/100000 progress=80.6% loss=0.068359 lr=9.941432e-07 elapsed=29:56:14 avg_step=1.34s recent_step=1.16s eta=07:12:20 +2026-08-26 09:51:56 | INFO | train rank=0 | step=80610/100000 progress=80.6% loss=0.054443 lr=9.931539e-07 elapsed=29:56:26 avg_step=1.34s recent_step=1.20s eta=07:12:07 +2026-08-26 09:52:15 | INFO | train rank=0 | step=80620/100000 progress=80.6% loss=0.043335 lr=9.921651e-07 elapsed=29:56:45 avg_step=1.34s recent_step=1.84s eta=07:11:54 +2026-08-26 09:52:27 | INFO | train rank=0 | step=80630/100000 progress=80.6% loss=0.069824 lr=9.911767e-07 elapsed=29:56:56 avg_step=1.34s recent_step=1.18s eta=07:11:41 +2026-08-26 09:52:39 | INFO | train rank=0 | step=80640/100000 progress=80.6% loss=0.046646 lr=9.901887e-07 elapsed=29:57:09 avg_step=1.34s recent_step=1.20s eta=07:11:27 +2026-08-26 09:52:50 | INFO | train rank=0 | step=80650/100000 progress=80.7% loss=0.043671 lr=9.892012e-07 elapsed=29:57:20 avg_step=1.34s recent_step=1.15s eta=07:11:13 +2026-08-26 09:53:05 | INFO | train rank=0 | step=80660/100000 progress=80.7% loss=0.024628 lr=9.882141e-07 elapsed=29:57:34 avg_step=1.34s recent_step=1.45s eta=07:11:00 +2026-08-26 09:53:16 | INFO | train rank=0 | step=80670/100000 progress=80.7% loss=0.038208 lr=9.872275e-07 elapsed=29:57:46 avg_step=1.34s recent_step=1.13s eta=07:10:46 +2026-08-26 09:53:32 | INFO | train rank=0 | step=80680/100000 progress=80.7% loss=0.035004 lr=9.862413e-07 elapsed=29:58:02 avg_step=1.34s recent_step=1.59s eta=07:10:33 +2026-08-26 09:53:44 | INFO | train rank=0 | step=80690/100000 progress=80.7% loss=0.045097 lr=9.852555e-07 elapsed=29:58:14 avg_step=1.34s recent_step=1.19s eta=07:10:20 +2026-08-26 09:54:01 | INFO | train rank=0 | step=80700/100000 progress=80.7% loss=0.037392 lr=9.842702e-07 elapsed=29:58:31 avg_step=1.34s recent_step=1.70s eta=07:10:07 +2026-08-26 09:54:13 | INFO | train rank=0 | step=80710/100000 progress=80.7% loss=0.051270 lr=9.832853e-07 elapsed=29:58:42 avg_step=1.34s recent_step=1.19s eta=07:09:53 +2026-08-26 09:54:29 | INFO | train rank=0 | step=80720/100000 progress=80.7% loss=0.041618 lr=9.823008e-07 elapsed=29:58:59 avg_step=1.34s recent_step=1.64s eta=07:09:41 +2026-08-26 09:54:40 | INFO | train rank=0 | step=80730/100000 progress=80.7% loss=0.046402 lr=9.813168e-07 elapsed=29:59:10 avg_step=1.34s recent_step=1.16s eta=07:09:27 +2026-08-26 09:54:52 | INFO | train rank=0 | step=80740/100000 progress=80.7% loss=0.032806 lr=9.803332e-07 elapsed=29:59:22 avg_step=1.34s recent_step=1.18s eta=07:09:13 +2026-08-26 09:55:04 | INFO | train rank=0 | step=80750/100000 progress=80.8% loss=0.040970 lr=9.793501e-07 elapsed=29:59:34 avg_step=1.34s recent_step=1.13s eta=07:08:59 +2026-08-26 09:55:17 | INFO | train rank=0 | step=80760/100000 progress=80.8% loss=0.034561 lr=9.783674e-07 elapsed=29:59:47 avg_step=1.34s recent_step=1.39s eta=07:08:46 +2026-08-26 09:55:30 | INFO | train rank=0 | step=80770/100000 progress=80.8% loss=0.035309 lr=9.773852e-07 elapsed=30:00:00 avg_step=1.34s recent_step=1.25s eta=07:08:33 +2026-08-26 09:55:44 | INFO | train rank=0 | step=80780/100000 progress=80.8% loss=0.044952 lr=9.764034e-07 elapsed=30:00:14 avg_step=1.34s recent_step=1.40s eta=07:08:19 +2026-08-26 09:55:58 | INFO | train rank=0 | step=80790/100000 progress=80.8% loss=0.049347 lr=9.754220e-07 elapsed=30:00:28 avg_step=1.34s recent_step=1.42s eta=07:08:06 +2026-08-26 09:56:18 | INFO | train rank=0 | step=80800/100000 progress=80.8% loss=0.037399 lr=9.744411e-07 elapsed=30:00:48 avg_step=1.34s recent_step=1.94s eta=07:07:54 +2026-08-26 09:56:29 | INFO | train rank=0 | step=80810/100000 progress=80.8% loss=0.037735 lr=9.734606e-07 elapsed=30:00:59 avg_step=1.34s recent_step=1.15s eta=07:07:40 +2026-08-26 09:56:41 | INFO | train rank=0 | step=80820/100000 progress=80.8% loss=0.106750 lr=9.724805e-07 elapsed=30:01:11 avg_step=1.34s recent_step=1.19s eta=07:07:27 +2026-08-26 09:56:57 | INFO | train rank=0 | step=80830/100000 progress=80.8% loss=0.030441 lr=9.715009e-07 elapsed=30:01:27 avg_step=1.34s recent_step=1.62s eta=07:07:14 +2026-08-26 09:57:09 | INFO | train rank=0 | step=80840/100000 progress=80.8% loss=0.037384 lr=9.705217e-07 elapsed=30:01:38 avg_step=1.34s recent_step=1.13s eta=07:07:00 +2026-08-26 09:57:22 | INFO | train rank=0 | step=80850/100000 progress=80.8% loss=0.041328 lr=9.695430e-07 elapsed=30:01:52 avg_step=1.34s recent_step=1.34s eta=07:06:47 +2026-08-26 09:57:33 | INFO | train rank=0 | step=80860/100000 progress=80.9% loss=0.021187 lr=9.685647e-07 elapsed=30:02:03 avg_step=1.34s recent_step=1.15s eta=07:06:33 +2026-08-26 09:57:48 | INFO | train rank=0 | step=80870/100000 progress=80.9% loss=0.051895 lr=9.675869e-07 elapsed=30:02:18 avg_step=1.34s recent_step=1.46s eta=07:06:20 +2026-08-26 09:58:00 | INFO | train rank=0 | step=80880/100000 progress=80.9% loss=0.047775 lr=9.666095e-07 elapsed=30:02:30 avg_step=1.34s recent_step=1.17s eta=07:06:06 +2026-08-26 09:58:14 | INFO | train rank=0 | step=80890/100000 progress=80.9% loss=0.070770 lr=9.656325e-07 elapsed=30:02:44 avg_step=1.34s recent_step=1.44s eta=07:05:53 +2026-08-26 09:58:27 | INFO | train rank=0 | step=80900/100000 progress=80.9% loss=0.028076 lr=9.646560e-07 elapsed=30:02:57 avg_step=1.34s recent_step=1.29s eta=07:05:40 +2026-08-26 09:58:43 | INFO | train rank=0 | step=80910/100000 progress=80.9% loss=0.022339 lr=9.636799e-07 elapsed=30:03:13 avg_step=1.34s recent_step=1.60s eta=07:05:27 +2026-08-26 09:58:55 | INFO | train rank=0 | step=80920/100000 progress=80.9% loss=0.044449 lr=9.627043e-07 elapsed=30:03:24 avg_step=1.34s recent_step=1.14s eta=07:05:13 +2026-08-26 09:59:06 | INFO | train rank=0 | step=80930/100000 progress=80.9% loss=0.039932 lr=9.617291e-07 elapsed=30:03:36 avg_step=1.34s recent_step=1.17s eta=07:04:59 +2026-08-26 09:59:20 | INFO | train rank=0 | step=80940/100000 progress=80.9% loss=0.068604 lr=9.607543e-07 elapsed=30:03:50 avg_step=1.34s recent_step=1.42s eta=07:04:46 +2026-08-26 09:59:32 | INFO | train rank=0 | step=80950/100000 progress=81.0% loss=0.034035 lr=9.597800e-07 elapsed=30:04:02 avg_step=1.34s recent_step=1.20s eta=07:04:32 +2026-08-26 09:59:48 | INFO | train rank=0 | step=80960/100000 progress=81.0% loss=0.042145 lr=9.588061e-07 elapsed=30:04:18 avg_step=1.34s recent_step=1.56s eta=07:04:19 +2026-08-26 10:00:00 | INFO | train rank=0 | step=80970/100000 progress=81.0% loss=0.027382 lr=9.578327e-07 elapsed=30:04:29 avg_step=1.34s recent_step=1.15s eta=07:04:06 +2026-08-26 10:00:12 | INFO | train rank=0 | step=80980/100000 progress=81.0% loss=0.032684 lr=9.568597e-07 elapsed=30:04:42 avg_step=1.34s recent_step=1.22s eta=07:03:52 +2026-08-26 10:00:23 | INFO | train rank=0 | step=80990/100000 progress=81.0% loss=0.031502 lr=9.558871e-07 elapsed=30:04:53 avg_step=1.34s recent_step=1.15s eta=07:03:38 +2026-08-26 10:00:37 | INFO | train rank=0 | step=81000/100000 progress=81.0% loss=0.050629 lr=9.549150e-07 elapsed=30:05:07 avg_step=1.34s recent_step=1.37s eta=07:03:25 +2026-08-26 10:00:49 | INFO | train rank=0 | step=81010/100000 progress=81.0% loss=0.061234 lr=9.539434e-07 elapsed=30:05:19 avg_step=1.34s recent_step=1.18s eta=07:03:11 +2026-08-26 10:01:04 | INFO | train rank=0 | step=81020/100000 progress=81.0% loss=0.050140 lr=9.529721e-07 elapsed=30:05:34 avg_step=1.34s recent_step=1.53s eta=07:02:58 +2026-08-26 10:01:24 | INFO | train rank=0 | step=81030/100000 progress=81.0% loss=0.041855 lr=9.520014e-07 elapsed=30:05:54 avg_step=1.34s recent_step=2.00s eta=07:02:46 +2026-08-26 10:01:37 | INFO | train rank=0 | step=81040/100000 progress=81.0% loss=0.036461 lr=9.510310e-07 elapsed=30:06:07 avg_step=1.34s recent_step=1.33s eta=07:02:33 +2026-08-26 10:01:49 | INFO | train rank=0 | step=81050/100000 progress=81.0% loss=0.053421 lr=9.500611e-07 elapsed=30:06:19 avg_step=1.34s recent_step=1.20s eta=07:02:19 +2026-08-26 10:02:03 | INFO | train rank=0 | step=81060/100000 progress=81.1% loss=0.027557 lr=9.490917e-07 elapsed=30:06:33 avg_step=1.34s recent_step=1.33s eta=07:02:06 +2026-08-26 10:02:18 | INFO | train rank=0 | step=81070/100000 progress=81.1% loss=0.039642 lr=9.481227e-07 elapsed=30:06:48 avg_step=1.34s recent_step=1.52s eta=07:01:53 +2026-08-26 10:02:30 | INFO | train rank=0 | step=81080/100000 progress=81.1% loss=0.029388 lr=9.471541e-07 elapsed=30:07:00 avg_step=1.34s recent_step=1.20s eta=07:01:39 +2026-08-26 10:02:47 | INFO | train rank=0 | step=81090/100000 progress=81.1% loss=0.083496 lr=9.461860e-07 elapsed=30:07:17 avg_step=1.34s recent_step=1.74s eta=07:01:27 +2026-08-26 10:03:01 | INFO | train rank=0 | step=81100/100000 progress=81.1% loss=0.035492 lr=9.452183e-07 elapsed=30:07:31 avg_step=1.34s recent_step=1.42s eta=07:01:14 +2026-08-26 10:03:15 | INFO | train rank=0 | step=81110/100000 progress=81.1% loss=0.044800 lr=9.442511e-07 elapsed=30:07:45 avg_step=1.34s recent_step=1.34s eta=07:01:00 +2026-08-26 10:03:26 | INFO | train rank=0 | step=81120/100000 progress=81.1% loss=0.053482 lr=9.432843e-07 elapsed=30:07:56 avg_step=1.34s recent_step=1.16s eta=07:00:47 +2026-08-26 10:03:39 | INFO | train rank=0 | step=81130/100000 progress=81.1% loss=0.068542 lr=9.423179e-07 elapsed=30:08:08 avg_step=1.34s recent_step=1.21s eta=07:00:33 +2026-08-26 10:03:53 | INFO | train rank=0 | step=81140/100000 progress=81.1% loss=0.040451 lr=9.413520e-07 elapsed=30:08:23 avg_step=1.34s recent_step=1.47s eta=07:00:20 +2026-08-26 10:04:06 | INFO | train rank=0 | step=81150/100000 progress=81.2% loss=0.051308 lr=9.403866e-07 elapsed=30:08:36 avg_step=1.34s recent_step=1.32s eta=07:00:06 +2026-08-26 10:04:20 | INFO | train rank=0 | step=81160/100000 progress=81.2% loss=0.052094 lr=9.394216e-07 elapsed=30:08:50 avg_step=1.34s recent_step=1.39s eta=06:59:53 +2026-08-26 10:04:32 | INFO | train rank=0 | step=81170/100000 progress=81.2% loss=0.044693 lr=9.384570e-07 elapsed=30:09:01 avg_step=1.34s recent_step=1.14s eta=06:59:39 +2026-08-26 10:04:44 | INFO | train rank=0 | step=81180/100000 progress=81.2% loss=0.037384 lr=9.374929e-07 elapsed=30:09:14 avg_step=1.34s recent_step=1.24s eta=06:59:26 +2026-08-26 10:04:56 | INFO | train rank=0 | step=81190/100000 progress=81.2% loss=0.039970 lr=9.365292e-07 elapsed=30:09:26 avg_step=1.34s recent_step=1.23s eta=06:59:12 +2026-08-26 10:05:12 | INFO | train rank=0 | step=81200/100000 progress=81.2% loss=0.068787 lr=9.355659e-07 elapsed=30:09:42 avg_step=1.34s recent_step=1.58s eta=06:58:59 +2026-08-26 10:05:27 | INFO | train rank=0 | step=81210/100000 progress=81.2% loss=0.043106 lr=9.346031e-07 elapsed=30:09:57 avg_step=1.34s recent_step=1.48s eta=06:58:46 +2026-08-26 10:05:43 | INFO | train rank=0 | step=81220/100000 progress=81.2% loss=0.055923 lr=9.336408e-07 elapsed=30:10:13 avg_step=1.34s recent_step=1.59s eta=06:58:33 +2026-08-26 10:05:53 | INFO | train rank=0 | step=81230/100000 progress=81.2% loss=0.042099 lr=9.326789e-07 elapsed=30:10:23 avg_step=1.34s recent_step=1.07s eta=06:58:19 +2026-08-26 10:06:06 | INFO | train rank=0 | step=81240/100000 progress=81.2% loss=0.063049 lr=9.317174e-07 elapsed=30:10:36 avg_step=1.34s recent_step=1.22s eta=06:58:06 +2026-08-26 10:06:17 | INFO | train rank=0 | step=81250/100000 progress=81.2% loss=0.037750 lr=9.307564e-07 elapsed=30:10:47 avg_step=1.34s recent_step=1.11s eta=06:57:52 +2026-08-26 10:06:28 | INFO | train rank=0 | step=81260/100000 progress=81.3% loss=0.024967 lr=9.297958e-07 elapsed=30:10:58 avg_step=1.34s recent_step=1.12s eta=06:57:38 +2026-08-26 10:06:46 | INFO | train rank=0 | step=81270/100000 progress=81.3% loss=0.053764 lr=9.288357e-07 elapsed=30:11:15 avg_step=1.34s recent_step=1.76s eta=06:57:26 +2026-08-26 10:06:57 | INFO | train rank=0 | step=81280/100000 progress=81.3% loss=0.081451 lr=9.278760e-07 elapsed=30:11:27 avg_step=1.34s recent_step=1.12s eta=06:57:12 +2026-08-26 10:07:11 | INFO | train rank=0 | step=81290/100000 progress=81.3% loss=0.043777 lr=9.269168e-07 elapsed=30:11:41 avg_step=1.34s recent_step=1.42s eta=06:56:59 +2026-08-26 10:07:27 | INFO | train rank=0 | step=81300/100000 progress=81.3% loss=0.061005 lr=9.259580e-07 elapsed=30:11:57 avg_step=1.34s recent_step=1.62s eta=06:56:46 +2026-08-26 10:07:39 | INFO | train rank=0 | step=81310/100000 progress=81.3% loss=0.045319 lr=9.249997e-07 elapsed=30:12:09 avg_step=1.34s recent_step=1.18s eta=06:56:32 +2026-08-26 10:07:53 | INFO | train rank=0 | step=81320/100000 progress=81.3% loss=0.038010 lr=9.240418e-07 elapsed=30:12:23 avg_step=1.34s recent_step=1.41s eta=06:56:19 +2026-08-26 10:08:07 | INFO | train rank=0 | step=81330/100000 progress=81.3% loss=0.035629 lr=9.230843e-07 elapsed=30:12:37 avg_step=1.34s recent_step=1.43s eta=06:56:06 +2026-08-26 10:08:20 | INFO | train rank=0 | step=81340/100000 progress=81.3% loss=0.029701 lr=9.221273e-07 elapsed=30:12:50 avg_step=1.34s recent_step=1.24s eta=06:55:52 +2026-08-26 10:08:36 | INFO | train rank=0 | step=81350/100000 progress=81.3% loss=0.046417 lr=9.211708e-07 elapsed=30:13:06 avg_step=1.34s recent_step=1.59s eta=06:55:39 +2026-08-26 10:08:49 | INFO | train rank=0 | step=81360/100000 progress=81.4% loss=0.088135 lr=9.202146e-07 elapsed=30:13:19 avg_step=1.34s recent_step=1.35s eta=06:55:26 +2026-08-26 10:09:05 | INFO | train rank=0 | step=81370/100000 progress=81.4% loss=0.095886 lr=9.192590e-07 elapsed=30:13:35 avg_step=1.34s recent_step=1.63s eta=06:55:13 +2026-08-26 10:09:19 | INFO | train rank=0 | step=81380/100000 progress=81.4% loss=0.050568 lr=9.183037e-07 elapsed=30:13:49 avg_step=1.34s recent_step=1.37s eta=06:55:00 +2026-08-26 10:09:34 | INFO | train rank=0 | step=81390/100000 progress=81.4% loss=0.045090 lr=9.173490e-07 elapsed=30:14:04 avg_step=1.34s recent_step=1.45s eta=06:54:47 +2026-08-26 10:09:49 | INFO | train rank=0 | step=81400/100000 progress=81.4% loss=0.034485 lr=9.163946e-07 elapsed=30:14:19 avg_step=1.34s recent_step=1.54s eta=06:54:34 +2026-08-26 10:10:00 | INFO | train rank=0 | step=81410/100000 progress=81.4% loss=0.042496 lr=9.154408e-07 elapsed=30:14:30 avg_step=1.34s recent_step=1.12s eta=06:54:20 +2026-08-26 10:10:15 | INFO | train rank=0 | step=81420/100000 progress=81.4% loss=0.031952 lr=9.144873e-07 elapsed=30:14:45 avg_step=1.34s recent_step=1.46s eta=06:54:07 +2026-08-26 10:10:27 | INFO | train rank=0 | step=81430/100000 progress=81.4% loss=0.056702 lr=9.135343e-07 elapsed=30:14:57 avg_step=1.34s recent_step=1.21s eta=06:53:53 +2026-08-26 10:10:43 | INFO | train rank=0 | step=81440/100000 progress=81.4% loss=0.050903 lr=9.125818e-07 elapsed=30:15:13 avg_step=1.34s recent_step=1.64s eta=06:53:41 +2026-08-26 10:10:57 | INFO | train rank=0 | step=81450/100000 progress=81.5% loss=0.045517 lr=9.116297e-07 elapsed=30:15:27 avg_step=1.34s recent_step=1.34s eta=06:53:27 +2026-08-26 10:11:08 | INFO | train rank=0 | step=81460/100000 progress=81.5% loss=0.030884 lr=9.106780e-07 elapsed=30:15:38 avg_step=1.34s recent_step=1.12s eta=06:53:14 +2026-08-26 10:11:23 | INFO | train rank=0 | step=81470/100000 progress=81.5% loss=0.066833 lr=9.097268e-07 elapsed=30:15:53 avg_step=1.34s recent_step=1.49s eta=06:53:00 +2026-08-26 10:11:36 | INFO | train rank=0 | step=81480/100000 progress=81.5% loss=0.036942 lr=9.087761e-07 elapsed=30:16:06 avg_step=1.34s recent_step=1.35s eta=06:52:47 +2026-08-26 10:11:48 | INFO | train rank=0 | step=81490/100000 progress=81.5% loss=0.034821 lr=9.078258e-07 elapsed=30:16:18 avg_step=1.34s recent_step=1.16s eta=06:52:33 +2026-08-26 10:12:00 | INFO | train rank=0 | step=81500/100000 progress=81.5% loss=0.060638 lr=9.068759e-07 elapsed=30:16:29 avg_step=1.34s recent_step=1.15s eta=06:52:20 +2026-08-26 10:12:11 | INFO | train rank=0 | step=81510/100000 progress=81.5% loss=0.039841 lr=9.059265e-07 elapsed=30:16:41 avg_step=1.34s recent_step=1.19s eta=06:52:06 +2026-08-26 10:12:30 | INFO | train rank=0 | step=81520/100000 progress=81.5% loss=0.076508 lr=9.049776e-07 elapsed=30:17:00 avg_step=1.34s recent_step=1.85s eta=06:51:54 +2026-08-26 10:12:42 | INFO | train rank=0 | step=81530/100000 progress=81.5% loss=0.053146 lr=9.040290e-07 elapsed=30:17:12 avg_step=1.34s recent_step=1.18s eta=06:51:40 +2026-08-26 10:12:55 | INFO | train rank=0 | step=81540/100000 progress=81.5% loss=0.062706 lr=9.030810e-07 elapsed=30:17:25 avg_step=1.34s recent_step=1.37s eta=06:51:27 +2026-08-26 10:13:07 | INFO | train rank=0 | step=81550/100000 progress=81.5% loss=0.053162 lr=9.021333e-07 elapsed=30:17:37 avg_step=1.34s recent_step=1.16s eta=06:51:13 +2026-08-26 10:13:19 | INFO | train rank=0 | step=81560/100000 progress=81.6% loss=0.044952 lr=9.011862e-07 elapsed=30:17:49 avg_step=1.34s recent_step=1.17s eta=06:50:59 +2026-08-26 10:13:33 | INFO | train rank=0 | step=81570/100000 progress=81.6% loss=0.037491 lr=9.002395e-07 elapsed=30:18:03 avg_step=1.34s recent_step=1.40s eta=06:50:46 +2026-08-26 10:13:44 | INFO | train rank=0 | step=81580/100000 progress=81.6% loss=0.057861 lr=8.992932e-07 elapsed=30:18:14 avg_step=1.34s recent_step=1.17s eta=06:50:32 +2026-08-26 10:13:59 | INFO | train rank=0 | step=81590/100000 progress=81.6% loss=0.037827 lr=8.983474e-07 elapsed=30:18:29 avg_step=1.34s recent_step=1.49s eta=06:50:19 +2026-08-26 10:14:11 | INFO | train rank=0 | step=81600/100000 progress=81.6% loss=0.037079 lr=8.974020e-07 elapsed=30:18:41 avg_step=1.34s recent_step=1.18s eta=06:50:05 +2026-08-26 10:14:22 | INFO | train rank=0 | step=81610/100000 progress=81.6% loss=0.044083 lr=8.964570e-07 elapsed=30:18:52 avg_step=1.34s recent_step=1.13s eta=06:49:52 +2026-08-26 10:14:36 | INFO | train rank=0 | step=81620/100000 progress=81.6% loss=0.044662 lr=8.955126e-07 elapsed=30:19:05 avg_step=1.34s recent_step=1.31s eta=06:49:38 +2026-08-26 10:14:54 | INFO | train rank=0 | step=81630/100000 progress=81.6% loss=0.042503 lr=8.945685e-07 elapsed=30:19:24 avg_step=1.34s recent_step=1.87s eta=06:49:26 +2026-08-26 10:15:06 | INFO | train rank=0 | step=81640/100000 progress=81.6% loss=0.041351 lr=8.936249e-07 elapsed=30:19:36 avg_step=1.34s recent_step=1.21s eta=06:49:12 +2026-08-26 10:15:18 | INFO | train rank=0 | step=81650/100000 progress=81.7% loss=0.047127 lr=8.926818e-07 elapsed=30:19:48 avg_step=1.34s recent_step=1.21s eta=06:48:59 +2026-08-26 10:15:33 | INFO | train rank=0 | step=81660/100000 progress=81.7% loss=0.039154 lr=8.917391e-07 elapsed=30:20:03 avg_step=1.34s recent_step=1.49s eta=06:48:46 +2026-08-26 10:15:48 | INFO | train rank=0 | step=81670/100000 progress=81.7% loss=0.046631 lr=8.907969e-07 elapsed=30:20:18 avg_step=1.34s recent_step=1.43s eta=06:48:32 +2026-08-26 10:15:59 | INFO | train rank=0 | step=81680/100000 progress=81.7% loss=0.060547 lr=8.898551e-07 elapsed=30:20:29 avg_step=1.34s recent_step=1.15s eta=06:48:19 +2026-08-26 10:16:15 | INFO | train rank=0 | step=81690/100000 progress=81.7% loss=0.042862 lr=8.889138e-07 elapsed=30:20:45 avg_step=1.34s recent_step=1.55s eta=06:48:06 +2026-08-26 10:16:27 | INFO | train rank=0 | step=81700/100000 progress=81.7% loss=0.032993 lr=8.879729e-07 elapsed=30:20:56 avg_step=1.34s recent_step=1.18s eta=06:47:52 +2026-08-26 10:16:40 | INFO | train rank=0 | step=81710/100000 progress=81.7% loss=0.056915 lr=8.870324e-07 elapsed=30:21:10 avg_step=1.34s recent_step=1.33s eta=06:47:39 +2026-08-26 10:16:53 | INFO | train rank=0 | step=81720/100000 progress=81.7% loss=0.033737 lr=8.860925e-07 elapsed=30:21:23 avg_step=1.34s recent_step=1.32s eta=06:47:25 +2026-08-26 10:17:07 | INFO | train rank=0 | step=81730/100000 progress=81.7% loss=0.067337 lr=8.851529e-07 elapsed=30:21:37 avg_step=1.34s recent_step=1.38s eta=06:47:12 +2026-08-26 10:17:21 | INFO | train rank=0 | step=81740/100000 progress=81.7% loss=0.065430 lr=8.842138e-07 elapsed=30:21:51 avg_step=1.34s recent_step=1.41s eta=06:46:59 +2026-08-26 10:17:33 | INFO | train rank=0 | step=81750/100000 progress=81.8% loss=0.031746 lr=8.832752e-07 elapsed=30:22:03 avg_step=1.34s recent_step=1.22s eta=06:46:45 +2026-08-26 10:17:44 | INFO | train rank=0 | step=81760/100000 progress=81.8% loss=0.030331 lr=8.823370e-07 elapsed=30:22:14 avg_step=1.34s recent_step=1.12s eta=06:46:31 +2026-08-26 10:18:00 | INFO | train rank=0 | step=81770/100000 progress=81.8% loss=0.045746 lr=8.813993e-07 elapsed=30:22:30 avg_step=1.34s recent_step=1.54s eta=06:46:18 +2026-08-26 10:18:14 | INFO | train rank=0 | step=81780/100000 progress=81.8% loss=0.041519 lr=8.804620e-07 elapsed=30:22:44 avg_step=1.34s recent_step=1.47s eta=06:46:05 +2026-08-26 10:18:26 | INFO | train rank=0 | step=81790/100000 progress=81.8% loss=0.039368 lr=8.795252e-07 elapsed=30:22:56 avg_step=1.34s recent_step=1.19s eta=06:45:52 +2026-08-26 10:18:38 | INFO | train rank=0 | step=81800/100000 progress=81.8% loss=0.050491 lr=8.785888e-07 elapsed=30:23:08 avg_step=1.34s recent_step=1.17s eta=06:45:38 +2026-08-26 10:18:54 | INFO | train rank=0 | step=81810/100000 progress=81.8% loss=0.112732 lr=8.776528e-07 elapsed=30:23:24 avg_step=1.34s recent_step=1.61s eta=06:45:25 +2026-08-26 10:19:06 | INFO | train rank=0 | step=81820/100000 progress=81.8% loss=0.043701 lr=8.767173e-07 elapsed=30:23:36 avg_step=1.34s recent_step=1.22s eta=06:45:11 +2026-08-26 10:19:20 | INFO | train rank=0 | step=81830/100000 progress=81.8% loss=0.029915 lr=8.757823e-07 elapsed=30:23:50 avg_step=1.34s recent_step=1.40s eta=06:44:58 +2026-08-26 10:19:32 | INFO | train rank=0 | step=81840/100000 progress=81.8% loss=0.039780 lr=8.748477e-07 elapsed=30:24:02 avg_step=1.34s recent_step=1.18s eta=06:44:44 +2026-08-26 10:19:46 | INFO | train rank=0 | step=81850/100000 progress=81.8% loss=0.038177 lr=8.739136e-07 elapsed=30:24:16 avg_step=1.34s recent_step=1.42s eta=06:44:31 +2026-08-26 10:19:58 | INFO | train rank=0 | step=81860/100000 progress=81.9% loss=0.050095 lr=8.729799e-07 elapsed=30:24:28 avg_step=1.34s recent_step=1.21s eta=06:44:18 +2026-08-26 10:20:13 | INFO | train rank=0 | step=81870/100000 progress=81.9% loss=0.029419 lr=8.720467e-07 elapsed=30:24:43 avg_step=1.34s recent_step=1.48s eta=06:44:04 +2026-08-26 10:20:26 | INFO | train rank=0 | step=81880/100000 progress=81.9% loss=0.032547 lr=8.711139e-07 elapsed=30:24:55 avg_step=1.34s recent_step=1.25s eta=06:43:51 +2026-08-26 10:20:37 | INFO | train rank=0 | step=81890/100000 progress=81.9% loss=0.043602 lr=8.701816e-07 elapsed=30:25:07 avg_step=1.34s recent_step=1.18s eta=06:43:37 +2026-08-26 10:20:51 | INFO | train rank=0 | step=81900/100000 progress=81.9% loss=0.059937 lr=8.692497e-07 elapsed=30:25:21 avg_step=1.34s recent_step=1.36s eta=06:43:24 +2026-08-26 10:21:05 | INFO | train rank=0 | step=81910/100000 progress=81.9% loss=0.047409 lr=8.683183e-07 elapsed=30:25:35 avg_step=1.34s recent_step=1.45s eta=06:43:11 +2026-08-26 10:21:19 | INFO | train rank=0 | step=81920/100000 progress=81.9% loss=0.048141 lr=8.673873e-07 elapsed=30:25:49 avg_step=1.34s recent_step=1.38s eta=06:42:57 +2026-08-26 10:21:35 | INFO | train rank=0 | step=81930/100000 progress=81.9% loss=0.045258 lr=8.664568e-07 elapsed=30:26:05 avg_step=1.34s recent_step=1.58s eta=06:42:45 +2026-08-26 10:21:50 | INFO | train rank=0 | step=81940/100000 progress=81.9% loss=0.063965 lr=8.655267e-07 elapsed=30:26:20 avg_step=1.34s recent_step=1.50s eta=06:42:32 +2026-08-26 10:22:05 | INFO | train rank=0 | step=81950/100000 progress=82.0% loss=0.045197 lr=8.645971e-07 elapsed=30:26:35 avg_step=1.34s recent_step=1.47s eta=06:42:18 +2026-08-26 10:22:16 | INFO | train rank=0 | step=81960/100000 progress=82.0% loss=0.035843 lr=8.636680e-07 elapsed=30:26:46 avg_step=1.34s recent_step=1.13s eta=06:42:05 +2026-08-26 10:22:30 | INFO | train rank=0 | step=81970/100000 progress=82.0% loss=0.047150 lr=8.627393e-07 elapsed=30:27:00 avg_step=1.34s recent_step=1.37s eta=06:41:51 +2026-08-26 10:22:43 | INFO | train rank=0 | step=81980/100000 progress=82.0% loss=0.040543 lr=8.618110e-07 elapsed=30:27:13 avg_step=1.34s recent_step=1.34s eta=06:41:38 +2026-08-26 10:22:55 | INFO | train rank=0 | step=81990/100000 progress=82.0% loss=0.036034 lr=8.608832e-07 elapsed=30:27:24 avg_step=1.34s recent_step=1.15s eta=06:41:24 +2026-08-26 10:23:06 | INFO | train rank=0 | step=82000/100000 progress=82.0% loss=0.053772 lr=8.599558e-07 elapsed=30:27:36 avg_step=1.34s recent_step=1.15s eta=06:41:10 +2026-08-26 10:23:22 | INFO | train rank=0 | step=82010/100000 progress=82.0% loss=0.051956 lr=8.590289e-07 elapsed=30:27:52 avg_step=1.34s recent_step=1.64s eta=06:40:58 +2026-08-26 10:23:35 | INFO | train rank=0 | step=82020/100000 progress=82.0% loss=0.048080 lr=8.581025e-07 elapsed=30:28:05 avg_step=1.34s recent_step=1.23s eta=06:40:44 +2026-08-26 10:23:50 | INFO | train rank=0 | step=82030/100000 progress=82.0% loss=0.053955 lr=8.571765e-07 elapsed=30:28:20 avg_step=1.34s recent_step=1.50s eta=06:40:31 +2026-08-26 10:24:04 | INFO | train rank=0 | step=82040/100000 progress=82.0% loss=0.033043 lr=8.562510e-07 elapsed=30:28:34 avg_step=1.34s recent_step=1.42s eta=06:40:18 +2026-08-26 10:24:16 | INFO | train rank=0 | step=82050/100000 progress=82.0% loss=0.044510 lr=8.553259e-07 elapsed=30:28:46 avg_step=1.34s recent_step=1.19s eta=06:40:04 +2026-08-26 10:24:28 | INFO | train rank=0 | step=82060/100000 progress=82.1% loss=0.042015 lr=8.544012e-07 elapsed=30:28:58 avg_step=1.34s recent_step=1.18s eta=06:39:50 +2026-08-26 10:24:41 | INFO | train rank=0 | step=82070/100000 progress=82.1% loss=0.055695 lr=8.534771e-07 elapsed=30:29:11 avg_step=1.34s recent_step=1.36s eta=06:39:37 +2026-08-26 10:24:55 | INFO | train rank=0 | step=82080/100000 progress=82.1% loss=0.032104 lr=8.525533e-07 elapsed=30:29:24 avg_step=1.34s recent_step=1.33s eta=06:39:24 +2026-08-26 10:25:10 | INFO | train rank=0 | step=82090/100000 progress=82.1% loss=0.053123 lr=8.516301e-07 elapsed=30:29:40 avg_step=1.34s recent_step=1.54s eta=06:39:11 +2026-08-26 10:25:23 | INFO | train rank=0 | step=82100/100000 progress=82.1% loss=0.046692 lr=8.507072e-07 elapsed=30:29:53 avg_step=1.34s recent_step=1.27s eta=06:38:57 +2026-08-26 10:25:36 | INFO | train rank=0 | step=82110/100000 progress=82.1% loss=0.082336 lr=8.497849e-07 elapsed=30:30:06 avg_step=1.34s recent_step=1.29s eta=06:38:44 +2026-08-26 10:25:50 | INFO | train rank=0 | step=82120/100000 progress=82.1% loss=0.052353 lr=8.488630e-07 elapsed=30:30:20 avg_step=1.34s recent_step=1.42s eta=06:38:31 +2026-08-26 10:26:03 | INFO | train rank=0 | step=82130/100000 progress=82.1% loss=0.066620 lr=8.479415e-07 elapsed=30:30:33 avg_step=1.34s recent_step=1.36s eta=06:38:17 +2026-08-26 10:26:18 | INFO | train rank=0 | step=82140/100000 progress=82.1% loss=0.065155 lr=8.470205e-07 elapsed=30:30:48 avg_step=1.34s recent_step=1.48s eta=06:38:04 +2026-08-26 10:26:30 | INFO | train rank=0 | step=82150/100000 progress=82.2% loss=0.070740 lr=8.461000e-07 elapsed=30:31:00 avg_step=1.34s recent_step=1.16s eta=06:37:51 +2026-08-26 10:26:43 | INFO | train rank=0 | step=82160/100000 progress=82.2% loss=0.043076 lr=8.451799e-07 elapsed=30:31:13 avg_step=1.34s recent_step=1.31s eta=06:37:37 +2026-08-26 10:26:54 | INFO | train rank=0 | step=82170/100000 progress=82.2% loss=0.056213 lr=8.442602e-07 elapsed=30:31:24 avg_step=1.34s recent_step=1.15s eta=06:37:23 +2026-08-26 10:27:10 | INFO | train rank=0 | step=82180/100000 progress=82.2% loss=0.052368 lr=8.433410e-07 elapsed=30:31:40 avg_step=1.34s recent_step=1.58s eta=06:37:10 +2026-08-26 10:27:21 | INFO | train rank=0 | step=82190/100000 progress=82.2% loss=0.054657 lr=8.424223e-07 elapsed=30:31:51 avg_step=1.34s recent_step=1.12s eta=06:36:57 +2026-08-26 10:27:36 | INFO | train rank=0 | step=82200/100000 progress=82.2% loss=0.028152 lr=8.415040e-07 elapsed=30:32:06 avg_step=1.34s recent_step=1.45s eta=06:36:44 +2026-08-26 10:27:48 | INFO | train rank=0 | step=82210/100000 progress=82.2% loss=0.046219 lr=8.405862e-07 elapsed=30:32:17 avg_step=1.34s recent_step=1.16s eta=06:36:30 +2026-08-26 10:28:04 | INFO | train rank=0 | step=82220/100000 progress=82.2% loss=0.074585 lr=8.396688e-07 elapsed=30:32:34 avg_step=1.34s recent_step=1.69s eta=06:36:17 +2026-08-26 10:28:17 | INFO | train rank=0 | step=82230/100000 progress=82.2% loss=0.040497 lr=8.387519e-07 elapsed=30:32:47 avg_step=1.34s recent_step=1.22s eta=06:36:04 +2026-08-26 10:28:29 | INFO | train rank=0 | step=82240/100000 progress=82.2% loss=0.042450 lr=8.378355e-07 elapsed=30:32:59 avg_step=1.34s recent_step=1.23s eta=06:35:50 +2026-08-26 10:28:42 | INFO | train rank=0 | step=82250/100000 progress=82.2% loss=0.045464 lr=8.369195e-07 elapsed=30:33:12 avg_step=1.34s recent_step=1.26s eta=06:35:36 +2026-08-26 10:28:55 | INFO | train rank=0 | step=82260/100000 progress=82.3% loss=0.050400 lr=8.360039e-07 elapsed=30:33:25 avg_step=1.34s recent_step=1.37s eta=06:35:23 +2026-08-26 10:29:09 | INFO | train rank=0 | step=82270/100000 progress=82.3% loss=0.044571 lr=8.350888e-07 elapsed=30:33:39 avg_step=1.34s recent_step=1.34s eta=06:35:10 +2026-08-26 10:29:20 | INFO | train rank=0 | step=82280/100000 progress=82.3% loss=0.040138 lr=8.341742e-07 elapsed=30:33:50 avg_step=1.34s recent_step=1.13s eta=06:34:56 +2026-08-26 10:29:35 | INFO | train rank=0 | step=82290/100000 progress=82.3% loss=0.043793 lr=8.332600e-07 elapsed=30:34:05 avg_step=1.34s recent_step=1.51s eta=06:34:43 +2026-08-26 10:29:49 | INFO | train rank=0 | step=82300/100000 progress=82.3% loss=0.033661 lr=8.323463e-07 elapsed=30:34:19 avg_step=1.34s recent_step=1.43s eta=06:34:30 +2026-08-26 10:30:03 | INFO | train rank=0 | step=82310/100000 progress=82.3% loss=0.054321 lr=8.314330e-07 elapsed=30:34:33 avg_step=1.34s recent_step=1.37s eta=06:34:16 +2026-08-26 10:30:18 | INFO | train rank=0 | step=82320/100000 progress=82.3% loss=0.050720 lr=8.305202e-07 elapsed=30:34:48 avg_step=1.34s recent_step=1.50s eta=06:34:03 +2026-08-26 10:30:31 | INFO | train rank=0 | step=82330/100000 progress=82.3% loss=0.054733 lr=8.296078e-07 elapsed=30:35:01 avg_step=1.34s recent_step=1.27s eta=06:33:50 +2026-08-26 10:30:46 | INFO | train rank=0 | step=82340/100000 progress=82.3% loss=0.051483 lr=8.286959e-07 elapsed=30:35:16 avg_step=1.34s recent_step=1.54s eta=06:33:37 +2026-08-26 10:30:58 | INFO | train rank=0 | step=82350/100000 progress=82.3% loss=0.033234 lr=8.277845e-07 elapsed=30:35:28 avg_step=1.34s recent_step=1.18s eta=06:33:23 +2026-08-26 10:31:15 | INFO | train rank=0 | step=82360/100000 progress=82.4% loss=0.078613 lr=8.268735e-07 elapsed=30:35:45 avg_step=1.34s recent_step=1.72s eta=06:33:11 +2026-08-26 10:31:29 | INFO | train rank=0 | step=82370/100000 progress=82.4% loss=0.046661 lr=8.259630e-07 elapsed=30:35:59 avg_step=1.34s recent_step=1.38s eta=06:32:57 +2026-08-26 10:31:40 | INFO | train rank=0 | step=82380/100000 progress=82.4% loss=0.022072 lr=8.250529e-07 elapsed=30:36:10 avg_step=1.34s recent_step=1.12s eta=06:32:44 +2026-08-26 10:31:52 | INFO | train rank=0 | step=82390/100000 progress=82.4% loss=0.037415 lr=8.241433e-07 elapsed=30:36:22 avg_step=1.34s recent_step=1.20s eta=06:32:30 +2026-08-26 10:32:04 | INFO | train rank=0 | step=82400/100000 progress=82.4% loss=0.092468 lr=8.232341e-07 elapsed=30:36:34 avg_step=1.34s recent_step=1.18s eta=06:32:16 +2026-08-26 10:32:20 | INFO | train rank=0 | step=82410/100000 progress=82.4% loss=0.038147 lr=8.223254e-07 elapsed=30:36:50 avg_step=1.34s recent_step=1.60s eta=06:32:03 +2026-08-26 10:32:36 | INFO | train rank=0 | step=82420/100000 progress=82.4% loss=0.040573 lr=8.214171e-07 elapsed=30:37:06 avg_step=1.34s recent_step=1.63s eta=06:31:51 +2026-08-26 10:32:52 | INFO | train rank=0 | step=82430/100000 progress=82.4% loss=0.043900 lr=8.205094e-07 elapsed=30:37:22 avg_step=1.34s recent_step=1.56s eta=06:31:38 +2026-08-26 10:33:04 | INFO | train rank=0 | step=82440/100000 progress=82.4% loss=0.050201 lr=8.196020e-07 elapsed=30:37:34 avg_step=1.34s recent_step=1.18s eta=06:31:24 +2026-08-26 10:33:16 | INFO | train rank=0 | step=82450/100000 progress=82.5% loss=0.046448 lr=8.186951e-07 elapsed=30:37:46 avg_step=1.34s recent_step=1.18s eta=06:31:10 +2026-08-26 10:33:27 | INFO | train rank=0 | step=82460/100000 progress=82.5% loss=0.033272 lr=8.177887e-07 elapsed=30:37:57 avg_step=1.34s recent_step=1.15s eta=06:30:57 +2026-08-26 10:33:44 | INFO | train rank=0 | step=82470/100000 progress=82.5% loss=0.031357 lr=8.168828e-07 elapsed=30:38:14 avg_step=1.34s recent_step=1.64s eta=06:30:44 +2026-08-26 10:34:02 | INFO | train rank=0 | step=82480/100000 progress=82.5% loss=0.102234 lr=8.159772e-07 elapsed=30:38:32 avg_step=1.34s recent_step=1.80s eta=06:30:31 +2026-08-26 10:34:13 | INFO | train rank=0 | step=82490/100000 progress=82.5% loss=0.065521 lr=8.150722e-07 elapsed=30:38:43 avg_step=1.34s recent_step=1.14s eta=06:30:18 +2026-08-26 10:34:25 | INFO | train rank=0 | step=82500/100000 progress=82.5% loss=0.051910 lr=8.141676e-07 elapsed=30:38:55 avg_step=1.34s recent_step=1.21s eta=06:30:04 +2026-08-26 10:34:37 | INFO | train rank=0 | step=82510/100000 progress=82.5% loss=0.046967 lr=8.132635e-07 elapsed=30:39:07 avg_step=1.34s recent_step=1.19s eta=06:29:50 +2026-08-26 10:34:49 | INFO | train rank=0 | step=82520/100000 progress=82.5% loss=0.049362 lr=8.123598e-07 elapsed=30:39:19 avg_step=1.34s recent_step=1.23s eta=06:29:37 +2026-08-26 10:35:02 | INFO | train rank=0 | step=82530/100000 progress=82.5% loss=0.076019 lr=8.114566e-07 elapsed=30:39:32 avg_step=1.34s recent_step=1.23s eta=06:29:23 +2026-08-26 10:35:14 | INFO | train rank=0 | step=82540/100000 progress=82.5% loss=0.062805 lr=8.105538e-07 elapsed=30:39:44 avg_step=1.34s recent_step=1.25s eta=06:29:10 +2026-08-26 10:35:30 | INFO | train rank=0 | step=82550/100000 progress=82.5% loss=0.041946 lr=8.096515e-07 elapsed=30:40:00 avg_step=1.34s recent_step=1.54s eta=06:28:57 +2026-08-26 10:35:46 | INFO | train rank=0 | step=82560/100000 progress=82.6% loss=0.035225 lr=8.087497e-07 elapsed=30:40:16 avg_step=1.34s recent_step=1.60s eta=06:28:44 +2026-08-26 10:36:02 | INFO | train rank=0 | step=82570/100000 progress=82.6% loss=0.055740 lr=8.078483e-07 elapsed=30:40:31 avg_step=1.34s recent_step=1.58s eta=06:28:31 +2026-08-26 10:36:13 | INFO | train rank=0 | step=82580/100000 progress=82.6% loss=0.065063 lr=8.069474e-07 elapsed=30:40:43 avg_step=1.34s recent_step=1.16s eta=06:28:17 +2026-08-26 10:36:31 | INFO | train rank=0 | step=82590/100000 progress=82.6% loss=0.070190 lr=8.060469e-07 elapsed=30:41:01 avg_step=1.34s recent_step=1.77s eta=06:28:05 +2026-08-26 10:36:43 | INFO | train rank=0 | step=82600/100000 progress=82.6% loss=0.049469 lr=8.051469e-07 elapsed=30:41:12 avg_step=1.34s recent_step=1.17s eta=06:27:51 +2026-08-26 10:36:54 | INFO | train rank=0 | step=82610/100000 progress=82.6% loss=0.066223 lr=8.042473e-07 elapsed=30:41:24 avg_step=1.34s recent_step=1.17s eta=06:27:37 +2026-08-26 10:37:08 | INFO | train rank=0 | step=82620/100000 progress=82.6% loss=0.058060 lr=8.033482e-07 elapsed=30:41:38 avg_step=1.34s recent_step=1.35s eta=06:27:24 +2026-08-26 10:37:23 | INFO | train rank=0 | step=82630/100000 progress=82.6% loss=0.039284 lr=8.024496e-07 elapsed=30:41:53 avg_step=1.34s recent_step=1.50s eta=06:27:11 +2026-08-26 10:37:34 | INFO | train rank=0 | step=82640/100000 progress=82.6% loss=0.030914 lr=8.015514e-07 elapsed=30:42:04 avg_step=1.34s recent_step=1.17s eta=06:26:57 +2026-08-26 10:37:47 | INFO | train rank=0 | step=82650/100000 progress=82.7% loss=0.117554 lr=8.006537e-07 elapsed=30:42:17 avg_step=1.34s recent_step=1.22s eta=06:26:44 +2026-08-26 10:38:00 | INFO | train rank=0 | step=82660/100000 progress=82.7% loss=0.048157 lr=7.997565e-07 elapsed=30:42:30 avg_step=1.34s recent_step=1.34s eta=06:26:30 +2026-08-26 10:38:19 | INFO | train rank=0 | step=82670/100000 progress=82.7% loss=0.076202 lr=7.988597e-07 elapsed=30:42:49 avg_step=1.34s recent_step=1.88s eta=06:26:18 +2026-08-26 10:38:31 | INFO | train rank=0 | step=82680/100000 progress=82.7% loss=0.033676 lr=7.979633e-07 elapsed=30:43:00 avg_step=1.34s recent_step=1.17s eta=06:26:04 +2026-08-26 10:38:43 | INFO | train rank=0 | step=82690/100000 progress=82.7% loss=0.041225 lr=7.970675e-07 elapsed=30:43:13 avg_step=1.34s recent_step=1.26s eta=06:25:51 +2026-08-26 10:38:55 | INFO | train rank=0 | step=82700/100000 progress=82.7% loss=0.026810 lr=7.961720e-07 elapsed=30:43:25 avg_step=1.34s recent_step=1.20s eta=06:25:37 +2026-08-26 10:39:08 | INFO | train rank=0 | step=82710/100000 progress=82.7% loss=0.034729 lr=7.952771e-07 elapsed=30:43:38 avg_step=1.34s recent_step=1.25s eta=06:25:23 +2026-08-26 10:39:20 | INFO | train rank=0 | step=82720/100000 progress=82.7% loss=0.046371 lr=7.943826e-07 elapsed=30:43:50 avg_step=1.34s recent_step=1.21s eta=06:25:10 +2026-08-26 10:39:34 | INFO | train rank=0 | step=82730/100000 progress=82.7% loss=0.057953 lr=7.934886e-07 elapsed=30:44:04 avg_step=1.34s recent_step=1.43s eta=06:24:57 +2026-08-26 10:39:46 | INFO | train rank=0 | step=82740/100000 progress=82.7% loss=0.029480 lr=7.925950e-07 elapsed=30:44:16 avg_step=1.34s recent_step=1.22s eta=06:24:43 +2026-08-26 10:40:01 | INFO | train rank=0 | step=82750/100000 progress=82.8% loss=0.031662 lr=7.917019e-07 elapsed=30:44:31 avg_step=1.34s recent_step=1.46s eta=06:24:30 +2026-08-26 10:40:12 | INFO | train rank=0 | step=82760/100000 progress=82.8% loss=0.046982 lr=7.908092e-07 elapsed=30:44:42 avg_step=1.34s recent_step=1.16s eta=06:24:16 +2026-08-26 10:40:27 | INFO | train rank=0 | step=82770/100000 progress=82.8% loss=0.050537 lr=7.899170e-07 elapsed=30:44:57 avg_step=1.34s recent_step=1.41s eta=06:24:03 +2026-08-26 10:40:39 | INFO | train rank=0 | step=82780/100000 progress=82.8% loss=0.046509 lr=7.890253e-07 elapsed=30:45:09 avg_step=1.34s recent_step=1.21s eta=06:23:49 +2026-08-26 10:40:54 | INFO | train rank=0 | step=82790/100000 progress=82.8% loss=0.086670 lr=7.881340e-07 elapsed=30:45:24 avg_step=1.34s recent_step=1.57s eta=06:23:36 +2026-08-26 10:41:09 | INFO | train rank=0 | step=82800/100000 progress=82.8% loss=0.082275 lr=7.872432e-07 elapsed=30:45:39 avg_step=1.34s recent_step=1.43s eta=06:23:23 +2026-08-26 10:41:21 | INFO | train rank=0 | step=82810/100000 progress=82.8% loss=0.040909 lr=7.863528e-07 elapsed=30:45:51 avg_step=1.34s recent_step=1.20s eta=06:23:10 +2026-08-26 10:41:35 | INFO | train rank=0 | step=82820/100000 progress=82.8% loss=0.049210 lr=7.854629e-07 elapsed=30:46:05 avg_step=1.34s recent_step=1.46s eta=06:22:56 +2026-08-26 10:41:50 | INFO | train rank=0 | step=82830/100000 progress=82.8% loss=0.056030 lr=7.845735e-07 elapsed=30:46:20 avg_step=1.34s recent_step=1.44s eta=06:22:43 +2026-08-26 10:42:02 | INFO | train rank=0 | step=82840/100000 progress=82.8% loss=0.050430 lr=7.836845e-07 elapsed=30:46:32 avg_step=1.34s recent_step=1.23s eta=06:22:30 +2026-08-26 10:42:14 | INFO | train rank=0 | step=82850/100000 progress=82.8% loss=0.065460 lr=7.827960e-07 elapsed=30:46:44 avg_step=1.34s recent_step=1.21s eta=06:22:16 +2026-08-26 10:42:29 | INFO | train rank=0 | step=82860/100000 progress=82.9% loss=0.134277 lr=7.819080e-07 elapsed=30:46:59 avg_step=1.34s recent_step=1.54s eta=06:22:03 +2026-08-26 10:42:45 | INFO | train rank=0 | step=82870/100000 progress=82.9% loss=0.045303 lr=7.810204e-07 elapsed=30:47:15 avg_step=1.34s recent_step=1.57s eta=06:21:50 +2026-08-26 10:42:57 | INFO | train rank=0 | step=82880/100000 progress=82.9% loss=0.053238 lr=7.801332e-07 elapsed=30:47:27 avg_step=1.34s recent_step=1.15s eta=06:21:36 +2026-08-26 10:43:12 | INFO | train rank=0 | step=82890/100000 progress=82.9% loss=0.042847 lr=7.792466e-07 elapsed=30:47:42 avg_step=1.34s recent_step=1.53s eta=06:21:24 +2026-08-26 10:43:24 | INFO | train rank=0 | step=82900/100000 progress=82.9% loss=0.050323 lr=7.783604e-07 elapsed=30:47:54 avg_step=1.34s recent_step=1.24s eta=06:21:10 +2026-08-26 10:43:38 | INFO | train rank=0 | step=82910/100000 progress=82.9% loss=0.095398 lr=7.774746e-07 elapsed=30:48:07 avg_step=1.34s recent_step=1.32s eta=06:20:57 +2026-08-26 10:43:50 | INFO | train rank=0 | step=82920/100000 progress=82.9% loss=0.036797 lr=7.765893e-07 elapsed=30:48:19 avg_step=1.34s recent_step=1.20s eta=06:20:43 +2026-08-26 10:44:01 | INFO | train rank=0 | step=82930/100000 progress=82.9% loss=0.037567 lr=7.757045e-07 elapsed=30:48:31 avg_step=1.34s recent_step=1.19s eta=06:20:29 +2026-08-26 10:44:19 | INFO | train rank=0 | step=82940/100000 progress=82.9% loss=0.048340 lr=7.748202e-07 elapsed=30:48:49 avg_step=1.34s recent_step=1.74s eta=06:20:17 +2026-08-26 10:44:30 | INFO | train rank=0 | step=82950/100000 progress=83.0% loss=0.033875 lr=7.739363e-07 elapsed=30:49:00 avg_step=1.34s recent_step=1.12s eta=06:20:03 +2026-08-26 10:44:44 | INFO | train rank=0 | step=82960/100000 progress=83.0% loss=0.052292 lr=7.730528e-07 elapsed=30:49:14 avg_step=1.34s recent_step=1.38s eta=06:19:50 +2026-08-26 10:44:59 | INFO | train rank=0 | step=82970/100000 progress=83.0% loss=0.052643 lr=7.721699e-07 elapsed=30:49:29 avg_step=1.34s recent_step=1.48s eta=06:19:36 +2026-08-26 10:45:13 | INFO | train rank=0 | step=82980/100000 progress=83.0% loss=0.041809 lr=7.712874e-07 elapsed=30:49:43 avg_step=1.34s recent_step=1.40s eta=06:19:23 +2026-08-26 10:45:28 | INFO | train rank=0 | step=82990/100000 progress=83.0% loss=0.065948 lr=7.704053e-07 elapsed=30:49:58 avg_step=1.34s recent_step=1.56s eta=06:19:10 +2026-08-26 10:45:41 | INFO | train rank=0 | step=83000/100000 progress=83.0% loss=0.052979 lr=7.695237e-07 elapsed=30:50:11 avg_step=1.34s recent_step=1.24s eta=06:18:57 +2026-08-26 10:45:55 | INFO | train rank=0 | step=83010/100000 progress=83.0% loss=0.040222 lr=7.686426e-07 elapsed=30:50:24 avg_step=1.34s recent_step=1.39s eta=06:18:43 +2026-08-26 10:46:07 | INFO | train rank=0 | step=83020/100000 progress=83.0% loss=0.050217 lr=7.677620e-07 elapsed=30:50:37 avg_step=1.34s recent_step=1.28s eta=06:18:30 +2026-08-26 10:46:20 | INFO | train rank=0 | step=83030/100000 progress=83.0% loss=0.051407 lr=7.668818e-07 elapsed=30:50:50 avg_step=1.34s recent_step=1.31s eta=06:18:16 +2026-08-26 10:46:33 | INFO | train rank=0 | step=83040/100000 progress=83.0% loss=0.042557 lr=7.660020e-07 elapsed=30:51:02 avg_step=1.34s recent_step=1.21s eta=06:18:03 +2026-08-26 10:46:44 | INFO | train rank=0 | step=83050/100000 progress=83.0% loss=0.045135 lr=7.651228e-07 elapsed=30:51:14 avg_step=1.34s recent_step=1.19s eta=06:17:49 +2026-08-26 10:46:55 | INFO | train rank=0 | step=83060/100000 progress=83.1% loss=0.038727 lr=7.642440e-07 elapsed=30:51:25 avg_step=1.34s recent_step=1.11s eta=06:17:35 +2026-08-26 10:47:10 | INFO | train rank=0 | step=83070/100000 progress=83.1% loss=0.040268 lr=7.633656e-07 elapsed=30:51:40 avg_step=1.34s recent_step=1.46s eta=06:17:22 +2026-08-26 10:47:22 | INFO | train rank=0 | step=83080/100000 progress=83.1% loss=0.046402 lr=7.624877e-07 elapsed=30:51:52 avg_step=1.34s recent_step=1.19s eta=06:17:09 +2026-08-26 10:47:34 | INFO | train rank=0 | step=83090/100000 progress=83.1% loss=0.060165 lr=7.616103e-07 elapsed=30:52:04 avg_step=1.34s recent_step=1.21s eta=06:16:55 +2026-08-26 10:47:47 | INFO | train rank=0 | step=83100/100000 progress=83.1% loss=0.053162 lr=7.607334e-07 elapsed=30:52:17 avg_step=1.34s recent_step=1.31s eta=06:16:41 +2026-08-26 10:47:58 | INFO | train rank=0 | step=83110/100000 progress=83.1% loss=0.054337 lr=7.598569e-07 elapsed=30:52:28 avg_step=1.34s recent_step=1.11s eta=06:16:28 +2026-08-26 10:48:13 | INFO | train rank=0 | step=83120/100000 progress=83.1% loss=0.055595 lr=7.589809e-07 elapsed=30:52:43 avg_step=1.34s recent_step=1.48s eta=06:16:15 +2026-08-26 10:48:30 | INFO | train rank=0 | step=83130/100000 progress=83.1% loss=0.049545 lr=7.581053e-07 elapsed=30:53:00 avg_step=1.34s recent_step=1.69s eta=06:16:02 +2026-08-26 10:48:42 | INFO | train rank=0 | step=83140/100000 progress=83.1% loss=0.048279 lr=7.572302e-07 elapsed=30:53:12 avg_step=1.34s recent_step=1.24s eta=06:15:48 +2026-08-26 10:48:58 | INFO | train rank=0 | step=83150/100000 progress=83.2% loss=0.062225 lr=7.563556e-07 elapsed=30:53:28 avg_step=1.34s recent_step=1.56s eta=06:15:35 +2026-08-26 10:49:12 | INFO | train rank=0 | step=83160/100000 progress=83.2% loss=0.030380 lr=7.554814e-07 elapsed=30:53:42 avg_step=1.34s recent_step=1.42s eta=06:15:22 +2026-08-26 10:49:26 | INFO | train rank=0 | step=83170/100000 progress=83.2% loss=0.052811 lr=7.546077e-07 elapsed=30:53:55 avg_step=1.34s recent_step=1.33s eta=06:15:09 +2026-08-26 10:49:38 | INFO | train rank=0 | step=83180/100000 progress=83.2% loss=0.042755 lr=7.537344e-07 elapsed=30:54:08 avg_step=1.34s recent_step=1.27s eta=06:14:55 +2026-08-26 10:49:51 | INFO | train rank=0 | step=83190/100000 progress=83.2% loss=0.040985 lr=7.528617e-07 elapsed=30:54:21 avg_step=1.34s recent_step=1.27s eta=06:14:42 +2026-08-26 10:50:05 | INFO | train rank=0 | step=83200/100000 progress=83.2% loss=0.040192 lr=7.519894e-07 elapsed=30:54:34 avg_step=1.34s recent_step=1.35s eta=06:14:28 +2026-08-26 10:50:20 | INFO | train rank=0 | step=83210/100000 progress=83.2% loss=0.054489 lr=7.511175e-07 elapsed=30:54:50 avg_step=1.34s recent_step=1.56s eta=06:14:16 +2026-08-26 10:50:32 | INFO | train rank=0 | step=83220/100000 progress=83.2% loss=0.024666 lr=7.502461e-07 elapsed=30:55:01 avg_step=1.34s recent_step=1.14s eta=06:14:02 +2026-08-26 10:50:46 | INFO | train rank=0 | step=83230/100000 progress=83.2% loss=0.044220 lr=7.493752e-07 elapsed=30:55:16 avg_step=1.34s recent_step=1.44s eta=06:13:49 +2026-08-26 10:50:57 | INFO | train rank=0 | step=83240/100000 progress=83.2% loss=0.078644 lr=7.485048e-07 elapsed=30:55:27 avg_step=1.34s recent_step=1.15s eta=06:13:35 +2026-08-26 10:51:09 | INFO | train rank=0 | step=83250/100000 progress=83.2% loss=0.041138 lr=7.476348e-07 elapsed=30:55:39 avg_step=1.34s recent_step=1.13s eta=06:13:21 +2026-08-26 10:51:27 | INFO | train rank=0 | step=83260/100000 progress=83.3% loss=0.051208 lr=7.467652e-07 elapsed=30:55:57 avg_step=1.34s recent_step=1.85s eta=06:13:09 +2026-08-26 10:51:40 | INFO | train rank=0 | step=83270/100000 progress=83.3% loss=0.034836 lr=7.458962e-07 elapsed=30:56:09 avg_step=1.34s recent_step=1.22s eta=06:12:55 +2026-08-26 10:51:53 | INFO | train rank=0 | step=83280/100000 progress=83.3% loss=0.049408 lr=7.450276e-07 elapsed=30:56:23 avg_step=1.34s recent_step=1.35s eta=06:12:42 +2026-08-26 10:52:09 | INFO | train rank=0 | step=83290/100000 progress=83.3% loss=0.051788 lr=7.441595e-07 elapsed=30:56:39 avg_step=1.34s recent_step=1.62s eta=06:12:29 +2026-08-26 10:52:21 | INFO | train rank=0 | step=83300/100000 progress=83.3% loss=0.057098 lr=7.432918e-07 elapsed=30:56:50 avg_step=1.34s recent_step=1.13s eta=06:12:15 +2026-08-26 10:52:37 | INFO | train rank=0 | step=83310/100000 progress=83.3% loss=0.030396 lr=7.424246e-07 elapsed=30:57:06 avg_step=1.34s recent_step=1.60s eta=06:12:02 +2026-08-26 10:52:48 | INFO | train rank=0 | step=83320/100000 progress=83.3% loss=0.032600 lr=7.415579e-07 elapsed=30:57:18 avg_step=1.34s recent_step=1.15s eta=06:11:49 +2026-08-26 10:53:00 | INFO | train rank=0 | step=83330/100000 progress=83.3% loss=0.049301 lr=7.406916e-07 elapsed=30:57:30 avg_step=1.34s recent_step=1.23s eta=06:11:35 +2026-08-26 10:53:12 | INFO | train rank=0 | step=83340/100000 progress=83.3% loss=0.048111 lr=7.398258e-07 elapsed=30:57:42 avg_step=1.34s recent_step=1.19s eta=06:11:21 +2026-08-26 10:53:29 | INFO | train rank=0 | step=83350/100000 progress=83.4% loss=0.052246 lr=7.389605e-07 elapsed=30:57:59 avg_step=1.34s recent_step=1.70s eta=06:11:09 +2026-08-26 10:53:42 | INFO | train rank=0 | step=83360/100000 progress=83.4% loss=0.044540 lr=7.380956e-07 elapsed=30:58:12 avg_step=1.34s recent_step=1.25s eta=06:10:55 +2026-08-26 10:53:56 | INFO | train rank=0 | step=83370/100000 progress=83.4% loss=0.046799 lr=7.372312e-07 elapsed=30:58:26 avg_step=1.34s recent_step=1.41s eta=06:10:42 +2026-08-26 10:54:11 | INFO | train rank=0 | step=83380/100000 progress=83.4% loss=0.057053 lr=7.363673e-07 elapsed=30:58:41 avg_step=1.34s recent_step=1.57s eta=06:10:29 +2026-08-26 10:54:23 | INFO | train rank=0 | step=83390/100000 progress=83.4% loss=0.026726 lr=7.355038e-07 elapsed=30:58:53 avg_step=1.34s recent_step=1.19s eta=06:10:15 +2026-08-26 10:54:35 | INFO | train rank=0 | step=83400/100000 progress=83.4% loss=0.043839 lr=7.346408e-07 elapsed=30:59:05 avg_step=1.34s recent_step=1.20s eta=06:10:02 +2026-08-26 10:54:47 | INFO | train rank=0 | step=83410/100000 progress=83.4% loss=0.032082 lr=7.337783e-07 elapsed=30:59:17 avg_step=1.34s recent_step=1.15s eta=06:09:48 +2026-08-26 10:55:02 | INFO | train rank=0 | step=83420/100000 progress=83.4% loss=0.024734 lr=7.329162e-07 elapsed=30:59:32 avg_step=1.34s recent_step=1.49s eta=06:09:35 +2026-08-26 10:55:13 | INFO | train rank=0 | step=83430/100000 progress=83.4% loss=0.050735 lr=7.320546e-07 elapsed=30:59:43 avg_step=1.34s recent_step=1.14s eta=06:09:21 +2026-08-26 10:55:25 | INFO | train rank=0 | step=83440/100000 progress=83.4% loss=0.041443 lr=7.311934e-07 elapsed=30:59:55 avg_step=1.34s recent_step=1.20s eta=06:09:07 +2026-08-26 10:55:38 | INFO | train rank=0 | step=83450/100000 progress=83.5% loss=0.049332 lr=7.303328e-07 elapsed=31:00:08 avg_step=1.34s recent_step=1.29s eta=06:08:54 +2026-08-26 10:55:53 | INFO | train rank=0 | step=83460/100000 progress=83.5% loss=0.040924 lr=7.294726e-07 elapsed=31:00:23 avg_step=1.34s recent_step=1.51s eta=06:08:41 +2026-08-26 10:56:07 | INFO | train rank=0 | step=83470/100000 progress=83.5% loss=0.062744 lr=7.286128e-07 elapsed=31:00:37 avg_step=1.34s recent_step=1.42s eta=06:08:28 +2026-08-26 10:56:22 | INFO | train rank=0 | step=83480/100000 progress=83.5% loss=0.052216 lr=7.277536e-07 elapsed=31:00:52 avg_step=1.34s recent_step=1.47s eta=06:08:15 +2026-08-26 10:56:34 | INFO | train rank=0 | step=83490/100000 progress=83.5% loss=0.031342 lr=7.268948e-07 elapsed=31:01:04 avg_step=1.34s recent_step=1.18s eta=06:08:01 +2026-08-26 10:56:46 | INFO | train rank=0 | step=83500/100000 progress=83.5% loss=0.041073 lr=7.260364e-07 elapsed=31:01:16 avg_step=1.34s recent_step=1.19s eta=06:07:47 +2026-08-26 10:57:02 | INFO | train rank=0 | step=83510/100000 progress=83.5% loss=0.044662 lr=7.251786e-07 elapsed=31:01:32 avg_step=1.34s recent_step=1.63s eta=06:07:34 +2026-08-26 10:57:14 | INFO | train rank=0 | step=83520/100000 progress=83.5% loss=0.042603 lr=7.243212e-07 elapsed=31:01:44 avg_step=1.34s recent_step=1.17s eta=06:07:21 +2026-08-26 10:57:29 | INFO | train rank=0 | step=83530/100000 progress=83.5% loss=0.057419 lr=7.234642e-07 elapsed=31:01:59 avg_step=1.34s recent_step=1.50s eta=06:07:08 +2026-08-26 10:57:40 | INFO | train rank=0 | step=83540/100000 progress=83.5% loss=0.043655 lr=7.226078e-07 elapsed=31:02:10 avg_step=1.34s recent_step=1.15s eta=06:06:54 +2026-08-26 10:57:57 | INFO | train rank=0 | step=83550/100000 progress=83.5% loss=0.040771 lr=7.217518e-07 elapsed=31:02:27 avg_step=1.34s recent_step=1.63s eta=06:06:41 +2026-08-26 10:58:08 | INFO | train rank=0 | step=83560/100000 progress=83.6% loss=0.059784 lr=7.208963e-07 elapsed=31:02:38 avg_step=1.34s recent_step=1.10s eta=06:06:27 +2026-08-26 10:58:24 | INFO | train rank=0 | step=83570/100000 progress=83.6% loss=0.025864 lr=7.200412e-07 elapsed=31:02:54 avg_step=1.34s recent_step=1.63s eta=06:06:15 +2026-08-26 10:58:37 | INFO | train rank=0 | step=83580/100000 progress=83.6% loss=0.052666 lr=7.191866e-07 elapsed=31:03:07 avg_step=1.34s recent_step=1.29s eta=06:06:01 +2026-08-26 10:58:49 | INFO | train rank=0 | step=83590/100000 progress=83.6% loss=0.073273 lr=7.183325e-07 elapsed=31:03:18 avg_step=1.34s recent_step=1.15s eta=06:05:47 +2026-08-26 10:59:04 | INFO | train rank=0 | step=83600/100000 progress=83.6% loss=0.038956 lr=7.174788e-07 elapsed=31:03:34 avg_step=1.34s recent_step=1.53s eta=06:05:34 +2026-08-26 10:59:15 | INFO | train rank=0 | step=83610/100000 progress=83.6% loss=0.032448 lr=7.166256e-07 elapsed=31:03:45 avg_step=1.34s recent_step=1.09s eta=06:05:21 +2026-08-26 10:59:26 | INFO | train rank=0 | step=83620/100000 progress=83.6% loss=0.043297 lr=7.157729e-07 elapsed=31:03:56 avg_step=1.34s recent_step=1.16s eta=06:05:07 +2026-08-26 10:59:39 | INFO | train rank=0 | step=83630/100000 progress=83.6% loss=0.043106 lr=7.149207e-07 elapsed=31:04:09 avg_step=1.34s recent_step=1.24s eta=06:04:53 +2026-08-26 10:59:53 | INFO | train rank=0 | step=83640/100000 progress=83.6% loss=0.035614 lr=7.140689e-07 elapsed=31:04:23 avg_step=1.34s recent_step=1.40s eta=06:04:40 +2026-08-26 11:00:07 | INFO | train rank=0 | step=83650/100000 progress=83.7% loss=0.076263 lr=7.132176e-07 elapsed=31:04:37 avg_step=1.34s recent_step=1.43s eta=06:04:27 +2026-08-26 11:00:22 | INFO | train rank=0 | step=83660/100000 progress=83.7% loss=0.051147 lr=7.123667e-07 elapsed=31:04:52 avg_step=1.34s recent_step=1.47s eta=06:04:14 +2026-08-26 11:00:36 | INFO | train rank=0 | step=83670/100000 progress=83.7% loss=0.036484 lr=7.115163e-07 elapsed=31:05:06 avg_step=1.34s recent_step=1.46s eta=06:04:01 +2026-08-26 11:00:49 | INFO | train rank=0 | step=83680/100000 progress=83.7% loss=0.039314 lr=7.106664e-07 elapsed=31:05:19 avg_step=1.34s recent_step=1.23s eta=06:03:47 +2026-08-26 11:01:01 | INFO | train rank=0 | step=83690/100000 progress=83.7% loss=0.038818 lr=7.098170e-07 elapsed=31:05:31 avg_step=1.34s recent_step=1.22s eta=06:03:33 +2026-08-26 11:01:13 | INFO | train rank=0 | step=83700/100000 progress=83.7% loss=0.037842 lr=7.089680e-07 elapsed=31:05:43 avg_step=1.34s recent_step=1.20s eta=06:03:20 +2026-08-26 11:01:27 | INFO | train rank=0 | step=83710/100000 progress=83.7% loss=0.047874 lr=7.081195e-07 elapsed=31:05:57 avg_step=1.34s recent_step=1.40s eta=06:03:06 +2026-08-26 11:01:40 | INFO | train rank=0 | step=83720/100000 progress=83.7% loss=0.047821 lr=7.072715e-07 elapsed=31:06:10 avg_step=1.34s recent_step=1.28s eta=06:02:53 +2026-08-26 11:01:55 | INFO | train rank=0 | step=83730/100000 progress=83.7% loss=0.044090 lr=7.064239e-07 elapsed=31:06:25 avg_step=1.34s recent_step=1.51s eta=06:02:40 +2026-08-26 11:02:09 | INFO | train rank=0 | step=83740/100000 progress=83.7% loss=0.044388 lr=7.055768e-07 elapsed=31:06:39 avg_step=1.34s recent_step=1.42s eta=06:02:27 +2026-08-26 11:02:21 | INFO | train rank=0 | step=83750/100000 progress=83.8% loss=0.061035 lr=7.047302e-07 elapsed=31:06:51 avg_step=1.34s recent_step=1.15s eta=06:02:13 +2026-08-26 11:02:33 | INFO | train rank=0 | step=83760/100000 progress=83.8% loss=0.036438 lr=7.038841e-07 elapsed=31:07:03 avg_step=1.34s recent_step=1.26s eta=06:01:59 +2026-08-26 11:02:49 | INFO | train rank=0 | step=83770/100000 progress=83.8% loss=0.043892 lr=7.030384e-07 elapsed=31:07:19 avg_step=1.34s recent_step=1.58s eta=06:01:47 +2026-08-26 11:03:03 | INFO | train rank=0 | step=83780/100000 progress=83.8% loss=0.040771 lr=7.021932e-07 elapsed=31:07:33 avg_step=1.34s recent_step=1.35s eta=06:01:33 +2026-08-26 11:03:16 | INFO | train rank=0 | step=83790/100000 progress=83.8% loss=0.028610 lr=7.013484e-07 elapsed=31:07:45 avg_step=1.34s recent_step=1.29s eta=06:01:20 +2026-08-26 11:03:33 | INFO | train rank=0 | step=83800/100000 progress=83.8% loss=0.034286 lr=7.005042e-07 elapsed=31:08:03 avg_step=1.34s recent_step=1.73s eta=06:01:07 +2026-08-26 11:03:45 | INFO | train rank=0 | step=83810/100000 progress=83.8% loss=0.032661 lr=6.996604e-07 elapsed=31:08:15 avg_step=1.34s recent_step=1.18s eta=06:00:53 +2026-08-26 11:03:56 | INFO | train rank=0 | step=83820/100000 progress=83.8% loss=0.043945 lr=6.988170e-07 elapsed=31:08:26 avg_step=1.34s recent_step=1.13s eta=06:00:40 +2026-08-26 11:04:11 | INFO | train rank=0 | step=83830/100000 progress=83.8% loss=0.044296 lr=6.979742e-07 elapsed=31:08:41 avg_step=1.34s recent_step=1.49s eta=06:00:27 +2026-08-26 11:04:25 | INFO | train rank=0 | step=83840/100000 progress=83.8% loss=0.050926 lr=6.971318e-07 elapsed=31:08:55 avg_step=1.34s recent_step=1.40s eta=06:00:13 +2026-08-26 11:04:39 | INFO | train rank=0 | step=83850/100000 progress=83.9% loss=0.032150 lr=6.962899e-07 elapsed=31:09:09 avg_step=1.34s recent_step=1.42s eta=06:00:00 +2026-08-26 11:04:51 | INFO | train rank=0 | step=83860/100000 progress=83.9% loss=0.040665 lr=6.954484e-07 elapsed=31:09:21 avg_step=1.34s recent_step=1.18s eta=05:59:46 +2026-08-26 11:05:03 | INFO | train rank=0 | step=83870/100000 progress=83.9% loss=0.036804 lr=6.946074e-07 elapsed=31:09:33 avg_step=1.34s recent_step=1.25s eta=05:59:33 +2026-08-26 11:05:18 | INFO | train rank=0 | step=83880/100000 progress=83.9% loss=0.057755 lr=6.937669e-07 elapsed=31:09:48 avg_step=1.34s recent_step=1.43s eta=05:59:20 +2026-08-26 11:05:35 | INFO | train rank=0 | step=83890/100000 progress=83.9% loss=0.045792 lr=6.929269e-07 elapsed=31:10:05 avg_step=1.34s recent_step=1.77s eta=05:59:07 +2026-08-26 11:05:46 | INFO | train rank=0 | step=83900/100000 progress=83.9% loss=0.034317 lr=6.920873e-07 elapsed=31:10:16 avg_step=1.34s recent_step=1.11s eta=05:58:53 +2026-08-26 11:05:58 | INFO | train rank=0 | step=83910/100000 progress=83.9% loss=0.030266 lr=6.912482e-07 elapsed=31:10:28 avg_step=1.34s recent_step=1.15s eta=05:58:40 +2026-08-26 11:06:10 | INFO | train rank=0 | step=83920/100000 progress=83.9% loss=0.060760 lr=6.904096e-07 elapsed=31:10:40 avg_step=1.34s recent_step=1.20s eta=05:58:26 +2026-08-26 11:06:24 | INFO | train rank=0 | step=83930/100000 progress=83.9% loss=0.033554 lr=6.895715e-07 elapsed=31:10:54 avg_step=1.34s recent_step=1.42s eta=05:58:13 +2026-08-26 11:06:39 | INFO | train rank=0 | step=83940/100000 progress=83.9% loss=0.025185 lr=6.887338e-07 elapsed=31:11:09 avg_step=1.34s recent_step=1.44s eta=05:58:00 +2026-08-26 11:06:53 | INFO | train rank=0 | step=83950/100000 progress=84.0% loss=0.032139 lr=6.878966e-07 elapsed=31:11:23 avg_step=1.34s recent_step=1.45s eta=05:57:46 +2026-08-26 11:07:05 | INFO | train rank=0 | step=83960/100000 progress=84.0% loss=0.021019 lr=6.870598e-07 elapsed=31:11:34 avg_step=1.34s recent_step=1.14s eta=05:57:33 +2026-08-26 11:07:20 | INFO | train rank=0 | step=83970/100000 progress=84.0% loss=0.069122 lr=6.862236e-07 elapsed=31:11:50 avg_step=1.34s recent_step=1.55s eta=05:57:20 +2026-08-26 11:07:32 | INFO | train rank=0 | step=83980/100000 progress=84.0% loss=0.030891 lr=6.853878e-07 elapsed=31:12:02 avg_step=1.34s recent_step=1.22s eta=05:57:06 +2026-08-26 11:07:44 | INFO | train rank=0 | step=83990/100000 progress=84.0% loss=0.039139 lr=6.845524e-07 elapsed=31:12:14 avg_step=1.34s recent_step=1.16s eta=05:56:52 +2026-08-26 11:07:56 | INFO | train rank=0 | step=84000/100000 progress=84.0% loss=0.055573 lr=6.837176e-07 elapsed=31:12:26 avg_step=1.34s recent_step=1.19s eta=05:56:39 +2026-08-26 11:08:09 | INFO | train rank=0 | step=84010/100000 progress=84.0% loss=0.038147 lr=6.828832e-07 elapsed=31:12:39 avg_step=1.34s recent_step=1.32s eta=05:56:25 +2026-08-26 11:08:22 | INFO | train rank=0 | step=84020/100000 progress=84.0% loss=0.043243 lr=6.820493e-07 elapsed=31:12:52 avg_step=1.34s recent_step=1.34s eta=05:56:12 +2026-08-26 11:08:37 | INFO | train rank=0 | step=84030/100000 progress=84.0% loss=0.043613 lr=6.812159e-07 elapsed=31:13:07 avg_step=1.34s recent_step=1.48s eta=05:55:59 +2026-08-26 11:08:49 | INFO | train rank=0 | step=84040/100000 progress=84.0% loss=0.047836 lr=6.803829e-07 elapsed=31:13:19 avg_step=1.34s recent_step=1.17s eta=05:55:45 +2026-08-26 11:09:04 | INFO | train rank=0 | step=84050/100000 progress=84.0% loss=0.034752 lr=6.795504e-07 elapsed=31:13:34 avg_step=1.34s recent_step=1.55s eta=05:55:32 +2026-08-26 11:09:19 | INFO | train rank=0 | step=84060/100000 progress=84.1% loss=0.047318 lr=6.787184e-07 elapsed=31:13:49 avg_step=1.34s recent_step=1.45s eta=05:55:19 +2026-08-26 11:09:30 | INFO | train rank=0 | step=84070/100000 progress=84.1% loss=0.034676 lr=6.778869e-07 elapsed=31:14:00 avg_step=1.34s recent_step=1.11s eta=05:55:05 +2026-08-26 11:09:46 | INFO | train rank=0 | step=84080/100000 progress=84.1% loss=0.047501 lr=6.770558e-07 elapsed=31:14:16 avg_step=1.34s recent_step=1.62s eta=05:54:52 +2026-08-26 11:09:58 | INFO | train rank=0 | step=84090/100000 progress=84.1% loss=0.046234 lr=6.762252e-07 elapsed=31:14:28 avg_step=1.34s recent_step=1.17s eta=05:54:39 +2026-08-26 11:10:10 | INFO | train rank=0 | step=84100/100000 progress=84.1% loss=0.071960 lr=6.753951e-07 elapsed=31:14:39 avg_step=1.34s recent_step=1.17s eta=05:54:25 +2026-08-26 11:10:24 | INFO | train rank=0 | step=84110/100000 progress=84.1% loss=0.065582 lr=6.745654e-07 elapsed=31:14:54 avg_step=1.34s recent_step=1.49s eta=05:54:12 +2026-08-26 11:10:36 | INFO | train rank=0 | step=84120/100000 progress=84.1% loss=0.051117 lr=6.737362e-07 elapsed=31:15:06 avg_step=1.34s recent_step=1.15s eta=05:53:58 +2026-08-26 11:10:50 | INFO | train rank=0 | step=84130/100000 progress=84.1% loss=0.038994 lr=6.729075e-07 elapsed=31:15:20 avg_step=1.34s recent_step=1.39s eta=05:53:45 +2026-08-26 11:11:02 | INFO | train rank=0 | step=84140/100000 progress=84.1% loss=0.028481 lr=6.720793e-07 elapsed=31:15:32 avg_step=1.34s recent_step=1.24s eta=05:53:31 +2026-08-26 11:11:19 | INFO | train rank=0 | step=84150/100000 progress=84.2% loss=0.126892 lr=6.712515e-07 elapsed=31:15:49 avg_step=1.34s recent_step=1.71s eta=05:53:19 +2026-08-26 11:11:35 | INFO | train rank=0 | step=84160/100000 progress=84.2% loss=0.033691 lr=6.704242e-07 elapsed=31:16:05 avg_step=1.34s recent_step=1.57s eta=05:53:06 +2026-08-26 11:11:47 | INFO | train rank=0 | step=84170/100000 progress=84.2% loss=0.041183 lr=6.695974e-07 elapsed=31:16:16 avg_step=1.34s recent_step=1.16s eta=05:52:52 +2026-08-26 11:11:59 | INFO | train rank=0 | step=84180/100000 progress=84.2% loss=0.073059 lr=6.687711e-07 elapsed=31:16:29 avg_step=1.34s recent_step=1.26s eta=05:52:39 +2026-08-26 11:12:13 | INFO | train rank=0 | step=84190/100000 progress=84.2% loss=0.050644 lr=6.679452e-07 elapsed=31:16:43 avg_step=1.34s recent_step=1.43s eta=05:52:25 +2026-08-26 11:12:29 | INFO | train rank=0 | step=84200/100000 progress=84.2% loss=0.033237 lr=6.671198e-07 elapsed=31:16:59 avg_step=1.34s recent_step=1.55s eta=05:52:12 +2026-08-26 11:12:41 | INFO | train rank=0 | step=84210/100000 progress=84.2% loss=0.045166 lr=6.662949e-07 elapsed=31:17:11 avg_step=1.34s recent_step=1.20s eta=05:51:59 +2026-08-26 11:12:52 | INFO | train rank=0 | step=84220/100000 progress=84.2% loss=0.058243 lr=6.654705e-07 elapsed=31:17:22 avg_step=1.34s recent_step=1.16s eta=05:51:45 +2026-08-26 11:13:04 | INFO | train rank=0 | step=84230/100000 progress=84.2% loss=0.027985 lr=6.646465e-07 elapsed=31:17:34 avg_step=1.34s recent_step=1.20s eta=05:51:31 +2026-08-26 11:13:18 | INFO | train rank=0 | step=84240/100000 progress=84.2% loss=0.038849 lr=6.638230e-07 elapsed=31:17:48 avg_step=1.34s recent_step=1.35s eta=05:51:18 +2026-08-26 11:13:33 | INFO | train rank=0 | step=84250/100000 progress=84.2% loss=0.072144 lr=6.630000e-07 elapsed=31:18:03 avg_step=1.34s recent_step=1.53s eta=05:51:05 +2026-08-26 11:13:53 | INFO | train rank=0 | step=84260/100000 progress=84.3% loss=0.035492 lr=6.621774e-07 elapsed=31:18:23 avg_step=1.34s recent_step=1.97s eta=05:50:53 +2026-08-26 11:14:05 | INFO | train rank=0 | step=84270/100000 progress=84.3% loss=0.054169 lr=6.613554e-07 elapsed=31:18:35 avg_step=1.34s recent_step=1.19s eta=05:50:39 +2026-08-26 11:14:16 | INFO | train rank=0 | step=84280/100000 progress=84.3% loss=0.038132 lr=6.605338e-07 elapsed=31:18:46 avg_step=1.34s recent_step=1.13s eta=05:50:25 +2026-08-26 11:14:31 | INFO | train rank=0 | step=84290/100000 progress=84.3% loss=0.035233 lr=6.597126e-07 elapsed=31:19:01 avg_step=1.34s recent_step=1.54s eta=05:50:12 +2026-08-26 11:14:46 | INFO | train rank=0 | step=84300/100000 progress=84.3% loss=0.033463 lr=6.588920e-07 elapsed=31:19:15 avg_step=1.34s recent_step=1.41s eta=05:49:59 +2026-08-26 11:14:57 | INFO | train rank=0 | step=84310/100000 progress=84.3% loss=0.037941 lr=6.580718e-07 elapsed=31:19:27 avg_step=1.34s recent_step=1.12s eta=05:49:45 +2026-08-26 11:15:09 | INFO | train rank=0 | step=84320/100000 progress=84.3% loss=0.032593 lr=6.572521e-07 elapsed=31:19:39 avg_step=1.34s recent_step=1.20s eta=05:49:32 +2026-08-26 11:15:20 | INFO | train rank=0 | step=84330/100000 progress=84.3% loss=0.042679 lr=6.564329e-07 elapsed=31:19:50 avg_step=1.34s recent_step=1.12s eta=05:49:18 +2026-08-26 11:15:40 | INFO | train rank=0 | step=84340/100000 progress=84.3% loss=0.043320 lr=6.556141e-07 elapsed=31:20:10 avg_step=1.34s recent_step=2.03s eta=05:49:06 +2026-08-26 11:15:52 | INFO | train rank=0 | step=84350/100000 progress=84.4% loss=0.032822 lr=6.547958e-07 elapsed=31:20:21 avg_step=1.34s recent_step=1.13s eta=05:48:52 +2026-08-26 11:16:09 | INFO | train rank=0 | step=84360/100000 progress=84.4% loss=0.051872 lr=6.539780e-07 elapsed=31:20:39 avg_step=1.34s recent_step=1.74s eta=05:48:39 +2026-08-26 11:16:22 | INFO | train rank=0 | step=84370/100000 progress=84.4% loss=0.048462 lr=6.531607e-07 elapsed=31:20:52 avg_step=1.34s recent_step=1.34s eta=05:48:26 +2026-08-26 11:16:34 | INFO | train rank=0 | step=84380/100000 progress=84.4% loss=0.031746 lr=6.523439e-07 elapsed=31:21:04 avg_step=1.34s recent_step=1.14s eta=05:48:12 +2026-08-26 11:16:45 | INFO | train rank=0 | step=84390/100000 progress=84.4% loss=0.034393 lr=6.515275e-07 elapsed=31:21:15 avg_step=1.34s recent_step=1.12s eta=05:47:59 +2026-08-26 11:16:59 | INFO | train rank=0 | step=84400/100000 progress=84.4% loss=0.051430 lr=6.507116e-07 elapsed=31:21:29 avg_step=1.34s recent_step=1.38s eta=05:47:45 +2026-08-26 11:17:12 | INFO | train rank=0 | step=84410/100000 progress=84.4% loss=0.043037 lr=6.498962e-07 elapsed=31:21:42 avg_step=1.34s recent_step=1.30s eta=05:47:32 +2026-08-26 11:17:30 | INFO | train rank=0 | step=84420/100000 progress=84.4% loss=0.060852 lr=6.490812e-07 elapsed=31:22:00 avg_step=1.34s recent_step=1.83s eta=05:47:19 +2026-08-26 11:17:41 | INFO | train rank=0 | step=84430/100000 progress=84.4% loss=0.057526 lr=6.482668e-07 elapsed=31:22:11 avg_step=1.34s recent_step=1.10s eta=05:47:06 +2026-08-26 11:17:54 | INFO | train rank=0 | step=84440/100000 progress=84.4% loss=0.086487 lr=6.474528e-07 elapsed=31:22:24 avg_step=1.34s recent_step=1.30s eta=05:46:52 +2026-08-26 11:18:06 | INFO | train rank=0 | step=84450/100000 progress=84.5% loss=0.119080 lr=6.466392e-07 elapsed=31:22:36 avg_step=1.34s recent_step=1.19s eta=05:46:38 +2026-08-26 11:18:21 | INFO | train rank=0 | step=84460/100000 progress=84.5% loss=0.056366 lr=6.458262e-07 elapsed=31:22:51 avg_step=1.34s recent_step=1.45s eta=05:46:25 +2026-08-26 11:18:32 | INFO | train rank=0 | step=84470/100000 progress=84.5% loss=0.036514 lr=6.450136e-07 elapsed=31:23:02 avg_step=1.34s recent_step=1.18s eta=05:46:12 +2026-08-26 11:18:49 | INFO | train rank=0 | step=84480/100000 progress=84.5% loss=0.085632 lr=6.442015e-07 elapsed=31:23:19 avg_step=1.34s recent_step=1.66s eta=05:45:59 +2026-08-26 11:19:00 | INFO | train rank=0 | step=84490/100000 progress=84.5% loss=0.049026 lr=6.433899e-07 elapsed=31:23:30 avg_step=1.34s recent_step=1.12s eta=05:45:45 +2026-08-26 11:19:12 | INFO | train rank=0 | step=84500/100000 progress=84.5% loss=0.036041 lr=6.425788e-07 elapsed=31:23:42 avg_step=1.34s recent_step=1.17s eta=05:45:31 +2026-08-26 11:19:28 | INFO | train rank=0 | step=84510/100000 progress=84.5% loss=0.036209 lr=6.417681e-07 elapsed=31:23:58 avg_step=1.34s recent_step=1.64s eta=05:45:19 +2026-08-26 11:19:40 | INFO | train rank=0 | step=84520/100000 progress=84.5% loss=0.056122 lr=6.409579e-07 elapsed=31:24:10 avg_step=1.34s recent_step=1.17s eta=05:45:05 +2026-08-26 11:19:52 | INFO | train rank=0 | step=84530/100000 progress=84.5% loss=0.050430 lr=6.401482e-07 elapsed=31:24:22 avg_step=1.34s recent_step=1.23s eta=05:44:51 +2026-08-26 11:20:10 | INFO | train rank=0 | step=84540/100000 progress=84.5% loss=0.069305 lr=6.393390e-07 elapsed=31:24:40 avg_step=1.34s recent_step=1.81s eta=05:44:39 +2026-08-26 11:20:26 | INFO | train rank=0 | step=84550/100000 progress=84.5% loss=0.062698 lr=6.385302e-07 elapsed=31:24:56 avg_step=1.34s recent_step=1.55s eta=05:44:26 +2026-08-26 11:20:37 | INFO | train rank=0 | step=84560/100000 progress=84.6% loss=0.050560 lr=6.377220e-07 elapsed=31:25:07 avg_step=1.34s recent_step=1.14s eta=05:44:12 +2026-08-26 11:20:51 | INFO | train rank=0 | step=84570/100000 progress=84.6% loss=0.025543 lr=6.369142e-07 elapsed=31:25:21 avg_step=1.34s recent_step=1.36s eta=05:43:59 +2026-08-26 11:21:03 | INFO | train rank=0 | step=84580/100000 progress=84.6% loss=0.053558 lr=6.361068e-07 elapsed=31:25:33 avg_step=1.34s recent_step=1.26s eta=05:43:45 +2026-08-26 11:21:17 | INFO | train rank=0 | step=84590/100000 progress=84.6% loss=0.058975 lr=6.353000e-07 elapsed=31:25:46 avg_step=1.34s recent_step=1.30s eta=05:43:32 +2026-08-26 11:21:28 | INFO | train rank=0 | step=84600/100000 progress=84.6% loss=0.044617 lr=6.344936e-07 elapsed=31:25:58 avg_step=1.34s recent_step=1.12s eta=05:43:18 +2026-08-26 11:21:42 | INFO | train rank=0 | step=84610/100000 progress=84.6% loss=0.059601 lr=6.336877e-07 elapsed=31:26:12 avg_step=1.34s recent_step=1.40s eta=05:43:05 +2026-08-26 11:21:55 | INFO | train rank=0 | step=84620/100000 progress=84.6% loss=0.042847 lr=6.328823e-07 elapsed=31:26:25 avg_step=1.34s recent_step=1.36s eta=05:42:51 +2026-08-26 11:22:09 | INFO | train rank=0 | step=84630/100000 progress=84.6% loss=0.054428 lr=6.320774e-07 elapsed=31:26:39 avg_step=1.34s recent_step=1.34s eta=05:42:38 +2026-08-26 11:22:20 | INFO | train rank=0 | step=84640/100000 progress=84.6% loss=0.054611 lr=6.312729e-07 elapsed=31:26:50 avg_step=1.34s recent_step=1.16s eta=05:42:24 +2026-08-26 11:22:31 | INFO | train rank=0 | step=84650/100000 progress=84.7% loss=0.074493 lr=6.304689e-07 elapsed=31:27:01 avg_step=1.34s recent_step=1.11s eta=05:42:11 +2026-08-26 11:22:46 | INFO | train rank=0 | step=84660/100000 progress=84.7% loss=0.098816 lr=6.296654e-07 elapsed=31:27:16 avg_step=1.34s recent_step=1.46s eta=05:41:57 +2026-08-26 11:22:58 | INFO | train rank=0 | step=84670/100000 progress=84.7% loss=0.034805 lr=6.288624e-07 elapsed=31:27:28 avg_step=1.34s recent_step=1.22s eta=05:41:44 +2026-08-26 11:23:10 | INFO | train rank=0 | step=84680/100000 progress=84.7% loss=0.045967 lr=6.280599e-07 elapsed=31:27:40 avg_step=1.34s recent_step=1.16s eta=05:41:30 +2026-08-26 11:23:22 | INFO | train rank=0 | step=84690/100000 progress=84.7% loss=0.057907 lr=6.272578e-07 elapsed=31:27:52 avg_step=1.34s recent_step=1.21s eta=05:41:16 +2026-08-26 11:23:37 | INFO | train rank=0 | step=84700/100000 progress=84.7% loss=0.064819 lr=6.264562e-07 elapsed=31:28:07 avg_step=1.34s recent_step=1.54s eta=05:41:03 +2026-08-26 11:23:49 | INFO | train rank=0 | step=84710/100000 progress=84.7% loss=0.055099 lr=6.256551e-07 elapsed=31:28:19 avg_step=1.34s recent_step=1.19s eta=05:40:50 +2026-08-26 11:24:03 | INFO | train rank=0 | step=84720/100000 progress=84.7% loss=0.068573 lr=6.248544e-07 elapsed=31:28:33 avg_step=1.34s recent_step=1.45s eta=05:40:37 +2026-08-26 11:24:15 | INFO | train rank=0 | step=84730/100000 progress=84.7% loss=0.038696 lr=6.240543e-07 elapsed=31:28:45 avg_step=1.34s recent_step=1.18s eta=05:40:23 +2026-08-26 11:24:35 | INFO | train rank=0 | step=84740/100000 progress=84.7% loss=0.046371 lr=6.232546e-07 elapsed=31:29:05 avg_step=1.34s recent_step=1.97s eta=05:40:11 +2026-08-26 11:24:48 | INFO | train rank=0 | step=84750/100000 progress=84.8% loss=0.040253 lr=6.224554e-07 elapsed=31:29:18 avg_step=1.34s recent_step=1.31s eta=05:39:57 +2026-08-26 11:25:00 | INFO | train rank=0 | step=84760/100000 progress=84.8% loss=0.086761 lr=6.216567e-07 elapsed=31:29:30 avg_step=1.34s recent_step=1.20s eta=05:39:44 +2026-08-26 11:25:15 | INFO | train rank=0 | step=84770/100000 progress=84.8% loss=0.042862 lr=6.208585e-07 elapsed=31:29:45 avg_step=1.34s recent_step=1.52s eta=05:39:31 +2026-08-26 11:25:32 | INFO | train rank=0 | step=84780/100000 progress=84.8% loss=0.052963 lr=6.200607e-07 elapsed=31:30:02 avg_step=1.34s recent_step=1.67s eta=05:39:18 +2026-08-26 11:25:47 | INFO | train rank=0 | step=84790/100000 progress=84.8% loss=0.040146 lr=6.192634e-07 elapsed=31:30:17 avg_step=1.34s recent_step=1.49s eta=05:39:05 +2026-08-26 11:26:01 | INFO | train rank=0 | step=84800/100000 progress=84.8% loss=0.062531 lr=6.184666e-07 elapsed=31:30:31 avg_step=1.34s recent_step=1.46s eta=05:38:52 +2026-08-26 11:26:16 | INFO | train rank=0 | step=84810/100000 progress=84.8% loss=0.038155 lr=6.176703e-07 elapsed=31:30:46 avg_step=1.34s recent_step=1.45s eta=05:38:38 +2026-08-26 11:26:27 | INFO | train rank=0 | step=84820/100000 progress=84.8% loss=0.070114 lr=6.168744e-07 elapsed=31:30:57 avg_step=1.34s recent_step=1.15s eta=05:38:25 +2026-08-26 11:26:39 | INFO | train rank=0 | step=84830/100000 progress=84.8% loss=0.056458 lr=6.160791e-07 elapsed=31:31:09 avg_step=1.34s recent_step=1.16s eta=05:38:11 +2026-08-26 11:26:51 | INFO | train rank=0 | step=84840/100000 progress=84.8% loss=0.048599 lr=6.152842e-07 elapsed=31:31:21 avg_step=1.34s recent_step=1.24s eta=05:37:57 +2026-08-26 11:27:07 | INFO | train rank=0 | step=84850/100000 progress=84.9% loss=0.043068 lr=6.144898e-07 elapsed=31:31:37 avg_step=1.34s recent_step=1.55s eta=05:37:45 +2026-08-26 11:27:20 | INFO | train rank=0 | step=84860/100000 progress=84.9% loss=0.040833 lr=6.136958e-07 elapsed=31:31:50 avg_step=1.34s recent_step=1.32s eta=05:37:31 +2026-08-26 11:27:34 | INFO | train rank=0 | step=84870/100000 progress=84.9% loss=0.067017 lr=6.129024e-07 elapsed=31:32:04 avg_step=1.34s recent_step=1.42s eta=05:37:18 +2026-08-26 11:27:48 | INFO | train rank=0 | step=84880/100000 progress=84.9% loss=0.057800 lr=6.121094e-07 elapsed=31:32:18 avg_step=1.34s recent_step=1.33s eta=05:37:04 +2026-08-26 11:28:05 | INFO | train rank=0 | step=84890/100000 progress=84.9% loss=0.082214 lr=6.113169e-07 elapsed=31:32:35 avg_step=1.34s recent_step=1.75s eta=05:36:52 +2026-08-26 11:28:17 | INFO | train rank=0 | step=84900/100000 progress=84.9% loss=0.029381 lr=6.105249e-07 elapsed=31:32:47 avg_step=1.34s recent_step=1.17s eta=05:36:38 +2026-08-26 11:28:29 | INFO | train rank=0 | step=84910/100000 progress=84.9% loss=0.041428 lr=6.097334e-07 elapsed=31:32:59 avg_step=1.34s recent_step=1.17s eta=05:36:25 +2026-08-26 11:28:41 | INFO | train rank=0 | step=84920/100000 progress=84.9% loss=0.049225 lr=6.089423e-07 elapsed=31:33:11 avg_step=1.34s recent_step=1.25s eta=05:36:11 +2026-08-26 11:28:52 | INFO | train rank=0 | step=84930/100000 progress=84.9% loss=0.067596 lr=6.081518e-07 elapsed=31:33:22 avg_step=1.34s recent_step=1.13s eta=05:35:57 +2026-08-26 11:29:09 | INFO | train rank=0 | step=84940/100000 progress=84.9% loss=0.042999 lr=6.073617e-07 elapsed=31:33:39 avg_step=1.34s recent_step=1.64s eta=05:35:44 +2026-08-26 11:29:20 | INFO | train rank=0 | step=84950/100000 progress=85.0% loss=0.061234 lr=6.065721e-07 elapsed=31:33:50 avg_step=1.34s recent_step=1.16s eta=05:35:31 +2026-08-26 11:29:36 | INFO | train rank=0 | step=84960/100000 progress=85.0% loss=0.070404 lr=6.057829e-07 elapsed=31:34:06 avg_step=1.34s recent_step=1.54s eta=05:35:18 +2026-08-26 11:29:47 | INFO | train rank=0 | step=84970/100000 progress=85.0% loss=0.037170 lr=6.049943e-07 elapsed=31:34:17 avg_step=1.34s recent_step=1.16s eta=05:35:04 +2026-08-26 11:29:59 | INFO | train rank=0 | step=84980/100000 progress=85.0% loss=0.046310 lr=6.042061e-07 elapsed=31:34:29 avg_step=1.34s recent_step=1.16s eta=05:34:50 +2026-08-26 11:30:17 | INFO | train rank=0 | step=84990/100000 progress=85.0% loss=0.033890 lr=6.034184e-07 elapsed=31:34:47 avg_step=1.34s recent_step=1.77s eta=05:34:38 +2026-08-26 11:30:29 | INFO | train rank=0 | step=85000/100000 progress=85.0% loss=0.053619 lr=6.026312e-07 elapsed=31:34:59 avg_step=1.34s recent_step=1.21s eta=05:34:24 +2026-08-26 11:30:40 | INFO | train rank=0 | step=85010/100000 progress=85.0% loss=0.059479 lr=6.018445e-07 elapsed=31:35:10 avg_step=1.34s recent_step=1.15s eta=05:34:10 +2026-08-26 11:30:52 | INFO | train rank=0 | step=85020/100000 progress=85.0% loss=0.030441 lr=6.010583e-07 elapsed=31:35:22 avg_step=1.34s recent_step=1.17s eta=05:33:57 +2026-08-26 11:31:08 | INFO | train rank=0 | step=85030/100000 progress=85.0% loss=0.034882 lr=6.002725e-07 elapsed=31:35:38 avg_step=1.34s recent_step=1.61s eta=05:33:44 +2026-08-26 11:31:20 | INFO | train rank=0 | step=85040/100000 progress=85.0% loss=0.042175 lr=5.994872e-07 elapsed=31:35:50 avg_step=1.34s recent_step=1.24s eta=05:33:30 +2026-08-26 11:31:32 | INFO | train rank=0 | step=85050/100000 progress=85.0% loss=0.078003 lr=5.987024e-07 elapsed=31:36:02 avg_step=1.34s recent_step=1.19s eta=05:33:17 +2026-08-26 11:31:48 | INFO | train rank=0 | step=85060/100000 progress=85.1% loss=0.100739 lr=5.979181e-07 elapsed=31:36:18 avg_step=1.34s recent_step=1.62s eta=05:33:04 +2026-08-26 11:32:02 | INFO | train rank=0 | step=85070/100000 progress=85.1% loss=0.034264 lr=5.971343e-07 elapsed=31:36:32 avg_step=1.34s recent_step=1.39s eta=05:32:50 +2026-08-26 11:32:14 | INFO | train rank=0 | step=85080/100000 progress=85.1% loss=0.047256 lr=5.963509e-07 elapsed=31:36:44 avg_step=1.34s recent_step=1.17s eta=05:32:37 +2026-08-26 11:32:28 | INFO | train rank=0 | step=85090/100000 progress=85.1% loss=0.031731 lr=5.955680e-07 elapsed=31:36:58 avg_step=1.34s recent_step=1.35s eta=05:32:23 +2026-08-26 11:32:44 | INFO | train rank=0 | step=85100/100000 progress=85.1% loss=0.037132 lr=5.947857e-07 elapsed=31:37:14 avg_step=1.34s recent_step=1.66s eta=05:32:11 +2026-08-26 11:32:56 | INFO | train rank=0 | step=85110/100000 progress=85.1% loss=0.060013 lr=5.940037e-07 elapsed=31:37:26 avg_step=1.34s recent_step=1.21s eta=05:31:57 +2026-08-26 11:33:11 | INFO | train rank=0 | step=85120/100000 progress=85.1% loss=0.035126 lr=5.932223e-07 elapsed=31:37:40 avg_step=1.34s recent_step=1.43s eta=05:31:44 +2026-08-26 11:33:26 | INFO | train rank=0 | step=85130/100000 progress=85.1% loss=0.064209 lr=5.924414e-07 elapsed=31:37:56 avg_step=1.34s recent_step=1.53s eta=05:31:31 +2026-08-26 11:33:41 | INFO | train rank=0 | step=85140/100000 progress=85.1% loss=0.071136 lr=5.916609e-07 elapsed=31:38:11 avg_step=1.34s recent_step=1.50s eta=05:31:18 +2026-08-26 11:33:52 | INFO | train rank=0 | step=85150/100000 progress=85.2% loss=0.049118 lr=5.908809e-07 elapsed=31:38:22 avg_step=1.34s recent_step=1.15s eta=05:31:04 +2026-08-26 11:34:04 | INFO | train rank=0 | step=85160/100000 progress=85.2% loss=0.033386 lr=5.901014e-07 elapsed=31:38:34 avg_step=1.34s recent_step=1.22s eta=05:30:50 +2026-08-26 11:34:20 | INFO | train rank=0 | step=85170/100000 progress=85.2% loss=0.041725 lr=5.893224e-07 elapsed=31:38:50 avg_step=1.34s recent_step=1.56s eta=05:30:37 +2026-08-26 11:34:32 | INFO | train rank=0 | step=85180/100000 progress=85.2% loss=0.048141 lr=5.885439e-07 elapsed=31:39:02 avg_step=1.34s recent_step=1.24s eta=05:30:24 +2026-08-26 11:34:44 | INFO | train rank=0 | step=85190/100000 progress=85.2% loss=0.070709 lr=5.877658e-07 elapsed=31:39:14 avg_step=1.34s recent_step=1.16s eta=05:30:10 +2026-08-26 11:34:58 | INFO | train rank=0 | step=85200/100000 progress=85.2% loss=0.058502 lr=5.869882e-07 elapsed=31:39:27 avg_step=1.34s recent_step=1.34s eta=05:29:57 +2026-08-26 11:35:10 | INFO | train rank=0 | step=85210/100000 progress=85.2% loss=0.071808 lr=5.862112e-07 elapsed=31:39:40 avg_step=1.34s recent_step=1.24s eta=05:29:43 +2026-08-26 11:35:26 | INFO | train rank=0 | step=85220/100000 progress=85.2% loss=0.047424 lr=5.854345e-07 elapsed=31:39:56 avg_step=1.34s recent_step=1.62s eta=05:29:30 +2026-08-26 11:35:38 | INFO | train rank=0 | step=85230/100000 progress=85.2% loss=0.045059 lr=5.846584e-07 elapsed=31:40:08 avg_step=1.34s recent_step=1.17s eta=05:29:17 +2026-08-26 11:35:49 | INFO | train rank=0 | step=85240/100000 progress=85.2% loss=0.096252 lr=5.838828e-07 elapsed=31:40:19 avg_step=1.34s recent_step=1.15s eta=05:29:03 +2026-08-26 11:36:02 | INFO | train rank=0 | step=85250/100000 progress=85.2% loss=0.050049 lr=5.831076e-07 elapsed=31:40:32 avg_step=1.34s recent_step=1.28s eta=05:28:49 +2026-08-26 11:36:14 | INFO | train rank=0 | step=85260/100000 progress=85.3% loss=0.061279 lr=5.823330e-07 elapsed=31:40:44 avg_step=1.34s recent_step=1.18s eta=05:28:36 +2026-08-26 11:36:26 | INFO | train rank=0 | step=85270/100000 progress=85.3% loss=0.057068 lr=5.815588e-07 elapsed=31:40:56 avg_step=1.34s recent_step=1.20s eta=05:28:22 +2026-08-26 11:36:45 | INFO | train rank=0 | step=85280/100000 progress=85.3% loss=0.066055 lr=5.807851e-07 elapsed=31:41:14 avg_step=1.34s recent_step=1.87s eta=05:28:10 +2026-08-26 11:36:56 | INFO | train rank=0 | step=85290/100000 progress=85.3% loss=0.041458 lr=5.800118e-07 elapsed=31:41:26 avg_step=1.34s recent_step=1.14s eta=05:27:56 +2026-08-26 11:37:09 | INFO | train rank=0 | step=85300/100000 progress=85.3% loss=0.048721 lr=5.792391e-07 elapsed=31:41:39 avg_step=1.34s recent_step=1.34s eta=05:27:43 +2026-08-26 11:37:21 | INFO | train rank=0 | step=85310/100000 progress=85.3% loss=0.064545 lr=5.784668e-07 elapsed=31:41:51 avg_step=1.34s recent_step=1.20s eta=05:27:29 +2026-08-26 11:37:40 | INFO | train rank=0 | step=85320/100000 progress=85.3% loss=0.040039 lr=5.776951e-07 elapsed=31:42:10 avg_step=1.34s recent_step=1.89s eta=05:27:17 +2026-08-26 11:37:52 | INFO | train rank=0 | step=85330/100000 progress=85.3% loss=0.049088 lr=5.769238e-07 elapsed=31:42:22 avg_step=1.34s recent_step=1.20s eta=05:27:03 +2026-08-26 11:38:07 | INFO | train rank=0 | step=85340/100000 progress=85.3% loss=0.042404 lr=5.761530e-07 elapsed=31:42:37 avg_step=1.34s recent_step=1.50s eta=05:26:50 +2026-08-26 11:38:20 | INFO | train rank=0 | step=85350/100000 progress=85.4% loss=0.034973 lr=5.753826e-07 elapsed=31:42:49 avg_step=1.34s recent_step=1.22s eta=05:26:36 +2026-08-26 11:38:34 | INFO | train rank=0 | step=85360/100000 progress=85.4% loss=0.040604 lr=5.746128e-07 elapsed=31:43:04 avg_step=1.34s recent_step=1.43s eta=05:26:23 +2026-08-26 11:38:47 | INFO | train rank=0 | step=85370/100000 progress=85.4% loss=0.047485 lr=5.738434e-07 elapsed=31:43:17 avg_step=1.34s recent_step=1.35s eta=05:26:10 +2026-08-26 11:39:03 | INFO | train rank=0 | step=85380/100000 progress=85.4% loss=0.048294 lr=5.730746e-07 elapsed=31:43:32 avg_step=1.34s recent_step=1.52s eta=05:25:57 +2026-08-26 11:39:14 | INFO | train rank=0 | step=85390/100000 progress=85.4% loss=0.061096 lr=5.723062e-07 elapsed=31:43:44 avg_step=1.34s recent_step=1.17s eta=05:25:43 +2026-08-26 11:39:31 | INFO | train rank=0 | step=85400/100000 progress=85.4% loss=0.037643 lr=5.715383e-07 elapsed=31:44:01 avg_step=1.34s recent_step=1.64s eta=05:25:30 +2026-08-26 11:39:42 | INFO | train rank=0 | step=85410/100000 progress=85.4% loss=0.039719 lr=5.707709e-07 elapsed=31:44:12 avg_step=1.34s recent_step=1.15s eta=05:25:16 +2026-08-26 11:39:57 | INFO | train rank=0 | step=85420/100000 progress=85.4% loss=0.037628 lr=5.700039e-07 elapsed=31:44:27 avg_step=1.34s recent_step=1.46s eta=05:25:03 +2026-08-26 11:40:15 | INFO | train rank=0 | step=85430/100000 progress=85.4% loss=0.041046 lr=5.692375e-07 elapsed=31:44:45 avg_step=1.34s recent_step=1.84s eta=05:24:51 +2026-08-26 11:40:27 | INFO | train rank=0 | step=85440/100000 progress=85.4% loss=0.031502 lr=5.684715e-07 elapsed=31:44:57 avg_step=1.34s recent_step=1.17s eta=05:24:37 +2026-08-26 11:40:39 | INFO | train rank=0 | step=85450/100000 progress=85.5% loss=0.051636 lr=5.677060e-07 elapsed=31:45:09 avg_step=1.34s recent_step=1.22s eta=05:24:24 +2026-08-26 11:40:51 | INFO | train rank=0 | step=85460/100000 progress=85.5% loss=0.041397 lr=5.669410e-07 elapsed=31:45:21 avg_step=1.34s recent_step=1.19s eta=05:24:10 +2026-08-26 11:41:03 | INFO | train rank=0 | step=85470/100000 progress=85.5% loss=0.035812 lr=5.661765e-07 elapsed=31:45:32 avg_step=1.34s recent_step=1.16s eta=05:23:56 +2026-08-26 11:41:18 | INFO | train rank=0 | step=85480/100000 progress=85.5% loss=0.039368 lr=5.654125e-07 elapsed=31:45:48 avg_step=1.34s recent_step=1.51s eta=05:23:43 +2026-08-26 11:41:29 | INFO | train rank=0 | step=85490/100000 progress=85.5% loss=0.047562 lr=5.646490e-07 elapsed=31:45:59 avg_step=1.34s recent_step=1.16s eta=05:23:29 +2026-08-26 11:41:47 | INFO | train rank=0 | step=85500/100000 progress=85.5% loss=0.032440 lr=5.638859e-07 elapsed=31:46:17 avg_step=1.34s recent_step=1.74s eta=05:23:17 +2026-08-26 11:42:00 | INFO | train rank=0 | step=85510/100000 progress=85.5% loss=0.075073 lr=5.631233e-07 elapsed=31:46:30 avg_step=1.34s recent_step=1.33s eta=05:23:03 +2026-08-26 11:42:13 | INFO | train rank=0 | step=85520/100000 progress=85.5% loss=0.032196 lr=5.623612e-07 elapsed=31:46:43 avg_step=1.34s recent_step=1.36s eta=05:22:50 +2026-08-26 11:42:28 | INFO | train rank=0 | step=85530/100000 progress=85.5% loss=0.058929 lr=5.615996e-07 elapsed=31:46:58 avg_step=1.34s recent_step=1.43s eta=05:22:37 +2026-08-26 11:42:39 | INFO | train rank=0 | step=85540/100000 progress=85.5% loss=0.044281 lr=5.608385e-07 elapsed=31:47:09 avg_step=1.34s recent_step=1.16s eta=05:22:23 +2026-08-26 11:42:56 | INFO | train rank=0 | step=85550/100000 progress=85.5% loss=0.035797 lr=5.600779e-07 elapsed=31:47:26 avg_step=1.34s recent_step=1.66s eta=05:22:10 +2026-08-26 11:43:08 | INFO | train rank=0 | step=85560/100000 progress=85.6% loss=0.045959 lr=5.593178e-07 elapsed=31:47:37 avg_step=1.34s recent_step=1.16s eta=05:21:57 +2026-08-26 11:43:20 | INFO | train rank=0 | step=85570/100000 progress=85.6% loss=0.032120 lr=5.585581e-07 elapsed=31:47:50 avg_step=1.34s recent_step=1.27s eta=05:21:43 +2026-08-26 11:43:34 | INFO | train rank=0 | step=85580/100000 progress=85.6% loss=0.046799 lr=5.577989e-07 elapsed=31:48:04 avg_step=1.34s recent_step=1.36s eta=05:21:30 +2026-08-26 11:43:47 | INFO | train rank=0 | step=85590/100000 progress=85.6% loss=0.042656 lr=5.570402e-07 elapsed=31:48:17 avg_step=1.34s recent_step=1.30s eta=05:21:16 +2026-08-26 11:44:00 | INFO | train rank=0 | step=85600/100000 progress=85.6% loss=0.072128 lr=5.562820e-07 elapsed=31:48:30 avg_step=1.34s recent_step=1.37s eta=05:21:03 +2026-08-26 11:44:13 | INFO | train rank=0 | step=85610/100000 progress=85.6% loss=0.033966 lr=5.555243e-07 elapsed=31:48:42 avg_step=1.34s recent_step=1.22s eta=05:20:49 +2026-08-26 11:44:29 | INFO | train rank=0 | step=85620/100000 progress=85.6% loss=0.046814 lr=5.547671e-07 elapsed=31:48:58 avg_step=1.34s recent_step=1.60s eta=05:20:36 +2026-08-26 11:44:40 | INFO | train rank=0 | step=85630/100000 progress=85.6% loss=0.041206 lr=5.540103e-07 elapsed=31:49:10 avg_step=1.34s recent_step=1.17s eta=05:20:23 +2026-08-26 11:44:54 | INFO | train rank=0 | step=85640/100000 progress=85.6% loss=0.042145 lr=5.532541e-07 elapsed=31:49:24 avg_step=1.34s recent_step=1.35s eta=05:20:09 +2026-08-26 11:45:09 | INFO | train rank=0 | step=85650/100000 progress=85.7% loss=0.051147 lr=5.524983e-07 elapsed=31:49:39 avg_step=1.34s recent_step=1.55s eta=05:19:56 +2026-08-26 11:45:23 | INFO | train rank=0 | step=85660/100000 progress=85.7% loss=0.021908 lr=5.517430e-07 elapsed=31:49:53 avg_step=1.34s recent_step=1.36s eta=05:19:43 +2026-08-26 11:45:35 | INFO | train rank=0 | step=85670/100000 progress=85.7% loss=0.074005 lr=5.509882e-07 elapsed=31:50:05 avg_step=1.34s recent_step=1.22s eta=05:19:30 +2026-08-26 11:45:47 | INFO | train rank=0 | step=85680/100000 progress=85.7% loss=0.036972 lr=5.502339e-07 elapsed=31:50:16 avg_step=1.34s recent_step=1.14s eta=05:19:16 +2026-08-26 11:45:58 | INFO | train rank=0 | step=85690/100000 progress=85.7% loss=0.046585 lr=5.494801e-07 elapsed=31:50:28 avg_step=1.34s recent_step=1.14s eta=05:19:02 +2026-08-26 11:46:13 | INFO | train rank=0 | step=85700/100000 progress=85.7% loss=0.050194 lr=5.487268e-07 elapsed=31:50:43 avg_step=1.34s recent_step=1.52s eta=05:18:49 +2026-08-26 11:46:25 | INFO | train rank=0 | step=85710/100000 progress=85.7% loss=0.034050 lr=5.479739e-07 elapsed=31:50:55 avg_step=1.34s recent_step=1.19s eta=05:18:35 +2026-08-26 11:46:40 | INFO | train rank=0 | step=85720/100000 progress=85.7% loss=0.069656 lr=5.472215e-07 elapsed=31:51:10 avg_step=1.34s recent_step=1.54s eta=05:18:22 +2026-08-26 11:46:56 | INFO | train rank=0 | step=85730/100000 progress=85.7% loss=0.054291 lr=5.464697e-07 elapsed=31:51:25 avg_step=1.34s recent_step=1.51s eta=05:18:09 +2026-08-26 11:47:07 | INFO | train rank=0 | step=85740/100000 progress=85.7% loss=0.043060 lr=5.457183e-07 elapsed=31:51:37 avg_step=1.34s recent_step=1.14s eta=05:17:56 +2026-08-26 11:47:22 | INFO | train rank=0 | step=85750/100000 progress=85.8% loss=0.059479 lr=5.449674e-07 elapsed=31:51:52 avg_step=1.34s recent_step=1.52s eta=05:17:43 +2026-08-26 11:47:36 | INFO | train rank=0 | step=85760/100000 progress=85.8% loss=0.030548 lr=5.442170e-07 elapsed=31:52:06 avg_step=1.34s recent_step=1.41s eta=05:17:29 +2026-08-26 11:47:51 | INFO | train rank=0 | step=85770/100000 progress=85.8% loss=0.051666 lr=5.434670e-07 elapsed=31:52:21 avg_step=1.34s recent_step=1.46s eta=05:17:16 +2026-08-26 11:48:04 | INFO | train rank=0 | step=85780/100000 progress=85.8% loss=0.067383 lr=5.427176e-07 elapsed=31:52:34 avg_step=1.34s recent_step=1.33s eta=05:17:03 +2026-08-26 11:48:18 | INFO | train rank=0 | step=85790/100000 progress=85.8% loss=0.040085 lr=5.419686e-07 elapsed=31:52:48 avg_step=1.34s recent_step=1.42s eta=05:16:49 +2026-08-26 11:48:33 | INFO | train rank=0 | step=85800/100000 progress=85.8% loss=0.045944 lr=5.412202e-07 elapsed=31:53:03 avg_step=1.34s recent_step=1.46s eta=05:16:36 +2026-08-26 11:48:44 | INFO | train rank=0 | step=85810/100000 progress=85.8% loss=0.029129 lr=5.404722e-07 elapsed=31:53:14 avg_step=1.34s recent_step=1.14s eta=05:16:23 +2026-08-26 11:48:57 | INFO | train rank=0 | step=85820/100000 progress=85.8% loss=0.038910 lr=5.397247e-07 elapsed=31:53:27 avg_step=1.34s recent_step=1.30s eta=05:16:09 +2026-08-26 11:49:12 | INFO | train rank=0 | step=85830/100000 progress=85.8% loss=0.031631 lr=5.389777e-07 elapsed=31:53:42 avg_step=1.34s recent_step=1.52s eta=05:15:56 +2026-08-26 11:49:26 | INFO | train rank=0 | step=85840/100000 progress=85.8% loss=0.041656 lr=5.382312e-07 elapsed=31:53:56 avg_step=1.34s recent_step=1.39s eta=05:15:43 +2026-08-26 11:49:38 | INFO | train rank=0 | step=85850/100000 progress=85.9% loss=0.036041 lr=5.374852e-07 elapsed=31:54:08 avg_step=1.34s recent_step=1.14s eta=05:15:29 +2026-08-26 11:49:52 | INFO | train rank=0 | step=85860/100000 progress=85.9% loss=0.047974 lr=5.367396e-07 elapsed=31:54:21 avg_step=1.34s recent_step=1.37s eta=05:15:16 +2026-08-26 11:50:07 | INFO | train rank=0 | step=85870/100000 progress=85.9% loss=0.073181 lr=5.359946e-07 elapsed=31:54:37 avg_step=1.34s recent_step=1.54s eta=05:15:03 +2026-08-26 11:50:18 | INFO | train rank=0 | step=85880/100000 progress=85.9% loss=0.049774 lr=5.352500e-07 elapsed=31:54:48 avg_step=1.34s recent_step=1.12s eta=05:14:49 +2026-08-26 11:50:33 | INFO | train rank=0 | step=85890/100000 progress=85.9% loss=0.057465 lr=5.345059e-07 elapsed=31:55:03 avg_step=1.34s recent_step=1.48s eta=05:14:36 +2026-08-26 11:50:44 | INFO | train rank=0 | step=85900/100000 progress=85.9% loss=0.047760 lr=5.337623e-07 elapsed=31:55:14 avg_step=1.34s recent_step=1.15s eta=05:14:22 +2026-08-26 11:50:58 | INFO | train rank=0 | step=85910/100000 progress=85.9% loss=0.030823 lr=5.330192e-07 elapsed=31:55:28 avg_step=1.34s recent_step=1.35s eta=05:14:09 +2026-08-26 11:51:14 | INFO | train rank=0 | step=85920/100000 progress=85.9% loss=0.041275 lr=5.322766e-07 elapsed=31:55:44 avg_step=1.34s recent_step=1.59s eta=05:13:56 +2026-08-26 11:51:25 | INFO | train rank=0 | step=85930/100000 progress=85.9% loss=0.028572 lr=5.315345e-07 elapsed=31:55:55 avg_step=1.34s recent_step=1.13s eta=05:13:42 +2026-08-26 11:51:39 | INFO | train rank=0 | step=85940/100000 progress=85.9% loss=0.047974 lr=5.307929e-07 elapsed=31:56:09 avg_step=1.34s recent_step=1.34s eta=05:13:29 +2026-08-26 11:51:53 | INFO | train rank=0 | step=85950/100000 progress=86.0% loss=0.042221 lr=5.300517e-07 elapsed=31:56:23 avg_step=1.34s recent_step=1.42s eta=05:13:15 +2026-08-26 11:52:09 | INFO | train rank=0 | step=85960/100000 progress=86.0% loss=0.032890 lr=5.293111e-07 elapsed=31:56:39 avg_step=1.34s recent_step=1.61s eta=05:13:03 +2026-08-26 11:52:20 | INFO | train rank=0 | step=85970/100000 progress=86.0% loss=0.049088 lr=5.285709e-07 elapsed=31:56:50 avg_step=1.34s recent_step=1.15s eta=05:12:49 +2026-08-26 11:52:32 | INFO | train rank=0 | step=85980/100000 progress=86.0% loss=0.089966 lr=5.278312e-07 elapsed=31:57:02 avg_step=1.34s recent_step=1.18s eta=05:12:35 +2026-08-26 11:52:44 | INFO | train rank=0 | step=85990/100000 progress=86.0% loss=0.040909 lr=5.270920e-07 elapsed=31:57:14 avg_step=1.34s recent_step=1.22s eta=05:12:22 +2026-08-26 11:52:56 | INFO | train rank=0 | step=86000/100000 progress=86.0% loss=0.047516 lr=5.263534e-07 elapsed=31:57:26 avg_step=1.34s recent_step=1.14s eta=05:12:08 +2026-08-26 11:53:08 | INFO | train rank=0 | step=86010/100000 progress=86.0% loss=0.064758 lr=5.256151e-07 elapsed=31:57:38 avg_step=1.34s recent_step=1.23s eta=05:11:54 +2026-08-26 11:53:20 | INFO | train rank=0 | step=86020/100000 progress=86.0% loss=0.041885 lr=5.248774e-07 elapsed=31:57:50 avg_step=1.34s recent_step=1.17s eta=05:11:41 +2026-08-26 11:53:35 | INFO | train rank=0 | step=86030/100000 progress=86.0% loss=0.031227 lr=5.241402e-07 elapsed=31:58:05 avg_step=1.34s recent_step=1.53s eta=05:11:28 +2026-08-26 11:53:48 | INFO | train rank=0 | step=86040/100000 progress=86.0% loss=0.061577 lr=5.234034e-07 elapsed=31:58:18 avg_step=1.34s recent_step=1.27s eta=05:11:14 +2026-08-26 11:54:02 | INFO | train rank=0 | step=86050/100000 progress=86.1% loss=0.034920 lr=5.226672e-07 elapsed=31:58:31 avg_step=1.34s recent_step=1.38s eta=05:11:01 +2026-08-26 11:54:14 | INFO | train rank=0 | step=86060/100000 progress=86.1% loss=0.036652 lr=5.219314e-07 elapsed=31:58:44 avg_step=1.34s recent_step=1.27s eta=05:10:47 +2026-08-26 11:54:30 | INFO | train rank=0 | step=86070/100000 progress=86.1% loss=0.066406 lr=5.211962e-07 elapsed=31:58:59 avg_step=1.34s recent_step=1.53s eta=05:10:34 +2026-08-26 11:54:41 | INFO | train rank=0 | step=86080/100000 progress=86.1% loss=0.023323 lr=5.204614e-07 elapsed=31:59:11 avg_step=1.34s recent_step=1.13s eta=05:10:21 +2026-08-26 11:54:52 | INFO | train rank=0 | step=86090/100000 progress=86.1% loss=0.056808 lr=5.197271e-07 elapsed=31:59:22 avg_step=1.34s recent_step=1.15s eta=05:10:07 +2026-08-26 11:55:06 | INFO | train rank=0 | step=86100/100000 progress=86.1% loss=0.060379 lr=5.189933e-07 elapsed=31:59:36 avg_step=1.34s recent_step=1.40s eta=05:09:54 +2026-08-26 11:55:20 | INFO | train rank=0 | step=86110/100000 progress=86.1% loss=0.057617 lr=5.182600e-07 elapsed=31:59:50 avg_step=1.34s recent_step=1.40s eta=05:09:40 +2026-08-26 11:55:36 | INFO | train rank=0 | step=86120/100000 progress=86.1% loss=0.049011 lr=5.175271e-07 elapsed=32:00:06 avg_step=1.34s recent_step=1.58s eta=05:09:27 +2026-08-26 11:55:51 | INFO | train rank=0 | step=86130/100000 progress=86.1% loss=0.036270 lr=5.167948e-07 elapsed=32:00:21 avg_step=1.34s recent_step=1.48s eta=05:09:14 +2026-08-26 11:56:05 | INFO | train rank=0 | step=86140/100000 progress=86.1% loss=0.042786 lr=5.160630e-07 elapsed=32:00:35 avg_step=1.34s recent_step=1.38s eta=05:09:01 +2026-08-26 11:56:18 | INFO | train rank=0 | step=86150/100000 progress=86.2% loss=0.032585 lr=5.153316e-07 elapsed=32:00:48 avg_step=1.34s recent_step=1.37s eta=05:08:48 +2026-08-26 11:56:30 | INFO | train rank=0 | step=86160/100000 progress=86.2% loss=0.031075 lr=5.146008e-07 elapsed=32:01:00 avg_step=1.34s recent_step=1.15s eta=05:08:34 +2026-08-26 11:56:45 | INFO | train rank=0 | step=86170/100000 progress=86.2% loss=0.029213 lr=5.138704e-07 elapsed=32:01:15 avg_step=1.34s recent_step=1.49s eta=05:08:21 +2026-08-26 11:57:02 | INFO | train rank=0 | step=86180/100000 progress=86.2% loss=0.035782 lr=5.131405e-07 elapsed=32:01:32 avg_step=1.34s recent_step=1.75s eta=05:08:08 +2026-08-26 11:57:16 | INFO | train rank=0 | step=86190/100000 progress=86.2% loss=0.049423 lr=5.124111e-07 elapsed=32:01:46 avg_step=1.34s recent_step=1.35s eta=05:07:55 +2026-08-26 11:57:31 | INFO | train rank=0 | step=86200/100000 progress=86.2% loss=0.069458 lr=5.116822e-07 elapsed=32:02:01 avg_step=1.34s recent_step=1.55s eta=05:07:42 +2026-08-26 11:57:44 | INFO | train rank=0 | step=86210/100000 progress=86.2% loss=0.059540 lr=5.109538e-07 elapsed=32:02:14 avg_step=1.34s recent_step=1.27s eta=05:07:28 +2026-08-26 11:57:56 | INFO | train rank=0 | step=86220/100000 progress=86.2% loss=0.043381 lr=5.102259e-07 elapsed=32:02:26 avg_step=1.34s recent_step=1.24s eta=05:07:15 +2026-08-26 11:58:08 | INFO | train rank=0 | step=86230/100000 progress=86.2% loss=0.030197 lr=5.094985e-07 elapsed=32:02:38 avg_step=1.34s recent_step=1.20s eta=05:07:01 +2026-08-26 11:58:20 | INFO | train rank=0 | step=86240/100000 progress=86.2% loss=0.058655 lr=5.087715e-07 elapsed=32:02:50 avg_step=1.34s recent_step=1.16s eta=05:06:47 +2026-08-26 11:58:34 | INFO | train rank=0 | step=86250/100000 progress=86.2% loss=0.035645 lr=5.080451e-07 elapsed=32:03:04 avg_step=1.34s recent_step=1.43s eta=05:06:34 +2026-08-26 11:58:48 | INFO | train rank=0 | step=86260/100000 progress=86.3% loss=0.058258 lr=5.073191e-07 elapsed=32:03:18 avg_step=1.34s recent_step=1.43s eta=05:06:21 +2026-08-26 11:59:00 | INFO | train rank=0 | step=86270/100000 progress=86.3% loss=0.029335 lr=5.065937e-07 elapsed=32:03:30 avg_step=1.34s recent_step=1.15s eta=05:06:07 +2026-08-26 11:59:12 | INFO | train rank=0 | step=86280/100000 progress=86.3% loss=0.050095 lr=5.058687e-07 elapsed=32:03:42 avg_step=1.34s recent_step=1.18s eta=05:05:54 +2026-08-26 11:59:24 | INFO | train rank=0 | step=86290/100000 progress=86.3% loss=0.042099 lr=5.051442e-07 elapsed=32:03:54 avg_step=1.34s recent_step=1.24s eta=05:05:40 +2026-08-26 11:59:39 | INFO | train rank=0 | step=86300/100000 progress=86.3% loss=0.064697 lr=5.044202e-07 elapsed=32:04:09 avg_step=1.34s recent_step=1.53s eta=05:05:27 +2026-08-26 11:59:54 | INFO | train rank=0 | step=86310/100000 progress=86.3% loss=0.048340 lr=5.036967e-07 elapsed=32:04:24 avg_step=1.34s recent_step=1.49s eta=05:05:14 +2026-08-26 12:00:06 | INFO | train rank=0 | step=86320/100000 progress=86.3% loss=0.057159 lr=5.029737e-07 elapsed=32:04:36 avg_step=1.34s recent_step=1.18s eta=05:05:00 +2026-08-26 12:00:21 | INFO | train rank=0 | step=86330/100000 progress=86.3% loss=0.045959 lr=5.022512e-07 elapsed=32:04:51 avg_step=1.34s recent_step=1.53s eta=05:04:47 +2026-08-26 12:00:37 | INFO | train rank=0 | step=86340/100000 progress=86.3% loss=0.022621 lr=5.015292e-07 elapsed=32:05:07 avg_step=1.34s recent_step=1.56s eta=05:04:34 +2026-08-26 12:00:49 | INFO | train rank=0 | step=86350/100000 progress=86.4% loss=0.045425 lr=5.008077e-07 elapsed=32:05:19 avg_step=1.34s recent_step=1.19s eta=05:04:20 +2026-08-26 12:01:00 | INFO | train rank=0 | step=86360/100000 progress=86.4% loss=0.052292 lr=5.000866e-07 elapsed=32:05:30 avg_step=1.34s recent_step=1.16s eta=05:04:07 +2026-08-26 12:01:12 | INFO | train rank=0 | step=86370/100000 progress=86.4% loss=0.035332 lr=4.993661e-07 elapsed=32:05:42 avg_step=1.34s recent_step=1.14s eta=05:03:53 +2026-08-26 12:01:29 | INFO | train rank=0 | step=86380/100000 progress=86.4% loss=0.060547 lr=4.986461e-07 elapsed=32:05:58 avg_step=1.34s recent_step=1.68s eta=05:03:40 +2026-08-26 12:01:40 | INFO | train rank=0 | step=86390/100000 progress=86.4% loss=0.074738 lr=4.979265e-07 elapsed=32:06:10 avg_step=1.34s recent_step=1.15s eta=05:03:27 +2026-08-26 12:01:55 | INFO | train rank=0 | step=86400/100000 progress=86.4% loss=0.033615 lr=4.972074e-07 elapsed=32:06:25 avg_step=1.34s recent_step=1.45s eta=05:03:13 +2026-08-26 12:02:06 | INFO | train rank=0 | step=86410/100000 progress=86.4% loss=0.021347 lr=4.964888e-07 elapsed=32:06:36 avg_step=1.34s recent_step=1.12s eta=05:03:00 +2026-08-26 12:02:20 | INFO | train rank=0 | step=86420/100000 progress=86.4% loss=0.042816 lr=4.957708e-07 elapsed=32:06:50 avg_step=1.34s recent_step=1.42s eta=05:02:46 +2026-08-26 12:02:34 | INFO | train rank=0 | step=86430/100000 progress=86.4% loss=0.057571 lr=4.950532e-07 elapsed=32:07:04 avg_step=1.34s recent_step=1.44s eta=05:02:33 +2026-08-26 12:02:48 | INFO | train rank=0 | step=86440/100000 progress=86.4% loss=0.059433 lr=4.943361e-07 elapsed=32:07:18 avg_step=1.34s recent_step=1.36s eta=05:02:20 +2026-08-26 12:02:59 | INFO | train rank=0 | step=86450/100000 progress=86.5% loss=0.034500 lr=4.936195e-07 elapsed=32:07:29 avg_step=1.34s recent_step=1.13s eta=05:02:06 +2026-08-26 12:03:11 | INFO | train rank=0 | step=86460/100000 progress=86.5% loss=0.042664 lr=4.929034e-07 elapsed=32:07:41 avg_step=1.34s recent_step=1.18s eta=05:01:53 +2026-08-26 12:03:26 | INFO | train rank=0 | step=86470/100000 progress=86.5% loss=0.051743 lr=4.921877e-07 elapsed=32:07:55 avg_step=1.34s recent_step=1.43s eta=05:01:39 +2026-08-26 12:03:40 | INFO | train rank=0 | step=86480/100000 progress=86.5% loss=0.060944 lr=4.914726e-07 elapsed=32:08:10 avg_step=1.34s recent_step=1.43s eta=05:01:26 +2026-08-26 12:03:52 | INFO | train rank=0 | step=86490/100000 progress=86.5% loss=0.053741 lr=4.907580e-07 elapsed=32:08:22 avg_step=1.34s recent_step=1.19s eta=05:01:13 +2026-08-26 12:04:08 | INFO | train rank=0 | step=86500/100000 progress=86.5% loss=0.044174 lr=4.900438e-07 elapsed=32:08:38 avg_step=1.34s recent_step=1.67s eta=05:01:00 +2026-08-26 12:04:23 | INFO | train rank=0 | step=86510/100000 progress=86.5% loss=0.043510 lr=4.893302e-07 elapsed=32:08:53 avg_step=1.34s recent_step=1.46s eta=05:00:46 +2026-08-26 12:04:39 | INFO | train rank=0 | step=86520/100000 progress=86.5% loss=0.039688 lr=4.886171e-07 elapsed=32:09:09 avg_step=1.34s recent_step=1.62s eta=05:00:34 +2026-08-26 12:04:50 | INFO | train rank=0 | step=86530/100000 progress=86.5% loss=0.040482 lr=4.879044e-07 elapsed=32:09:20 avg_step=1.34s recent_step=1.11s eta=05:00:20 +2026-08-26 12:05:07 | INFO | train rank=0 | step=86540/100000 progress=86.5% loss=0.038269 lr=4.871922e-07 elapsed=32:09:37 avg_step=1.34s recent_step=1.65s eta=05:00:07 +2026-08-26 12:05:20 | INFO | train rank=0 | step=86550/100000 progress=86.6% loss=0.066254 lr=4.864806e-07 elapsed=32:09:50 avg_step=1.34s recent_step=1.36s eta=04:59:54 +2026-08-26 12:05:32 | INFO | train rank=0 | step=86560/100000 progress=86.6% loss=0.046112 lr=4.857694e-07 elapsed=32:10:02 avg_step=1.34s recent_step=1.19s eta=04:59:40 +2026-08-26 12:05:44 | INFO | train rank=0 | step=86570/100000 progress=86.6% loss=0.037781 lr=4.850587e-07 elapsed=32:10:14 avg_step=1.34s recent_step=1.20s eta=04:59:26 +2026-08-26 12:05:57 | INFO | train rank=0 | step=86580/100000 progress=86.6% loss=0.044830 lr=4.843485e-07 elapsed=32:10:27 avg_step=1.34s recent_step=1.24s eta=04:59:13 +2026-08-26 12:06:12 | INFO | train rank=0 | step=86590/100000 progress=86.6% loss=0.036919 lr=4.836388e-07 elapsed=32:10:42 avg_step=1.34s recent_step=1.56s eta=04:59:00 +2026-08-26 12:06:24 | INFO | train rank=0 | step=86600/100000 progress=86.6% loss=0.041153 lr=4.829296e-07 elapsed=32:10:53 avg_step=1.34s recent_step=1.12s eta=04:58:46 +2026-08-26 12:06:37 | INFO | train rank=0 | step=86610/100000 progress=86.6% loss=0.049500 lr=4.822209e-07 elapsed=32:11:07 avg_step=1.34s recent_step=1.33s eta=04:58:33 +2026-08-26 12:06:54 | INFO | train rank=0 | step=86620/100000 progress=86.6% loss=0.063644 lr=4.815127e-07 elapsed=32:11:24 avg_step=1.34s recent_step=1.69s eta=04:58:20 +2026-08-26 12:07:11 | INFO | train rank=0 | step=86630/100000 progress=86.6% loss=0.036766 lr=4.808049e-07 elapsed=32:11:41 avg_step=1.34s recent_step=1.75s eta=04:58:07 +2026-08-26 12:07:22 | INFO | train rank=0 | step=86640/100000 progress=86.6% loss=0.029869 lr=4.800977e-07 elapsed=32:11:52 avg_step=1.34s recent_step=1.12s eta=04:57:53 +2026-08-26 12:07:34 | INFO | train rank=0 | step=86650/100000 progress=86.7% loss=0.030716 lr=4.793910e-07 elapsed=32:12:04 avg_step=1.34s recent_step=1.14s eta=04:57:40 +2026-08-26 12:07:45 | INFO | train rank=0 | step=86660/100000 progress=86.7% loss=0.058990 lr=4.786847e-07 elapsed=32:12:15 avg_step=1.34s recent_step=1.11s eta=04:57:26 +2026-08-26 12:07:57 | INFO | train rank=0 | step=86670/100000 progress=86.7% loss=0.057312 lr=4.779790e-07 elapsed=32:12:27 avg_step=1.34s recent_step=1.20s eta=04:57:12 +2026-08-26 12:08:08 | INFO | train rank=0 | step=86680/100000 progress=86.7% loss=0.052414 lr=4.772738e-07 elapsed=32:12:38 avg_step=1.34s recent_step=1.11s eta=04:56:59 +2026-08-26 12:08:19 | INFO | train rank=0 | step=86690/100000 progress=86.7% loss=0.080750 lr=4.765690e-07 elapsed=32:12:49 avg_step=1.34s recent_step=1.13s eta=04:56:45 +2026-08-26 12:08:32 | INFO | train rank=0 | step=86700/100000 progress=86.7% loss=0.044525 lr=4.758647e-07 elapsed=32:13:01 avg_step=1.34s recent_step=1.23s eta=04:56:31 +2026-08-26 12:08:45 | INFO | train rank=0 | step=86710/100000 progress=86.7% loss=0.040588 lr=4.751610e-07 elapsed=32:13:15 avg_step=1.34s recent_step=1.38s eta=04:56:18 +2026-08-26 12:08:57 | INFO | train rank=0 | step=86720/100000 progress=86.7% loss=0.035126 lr=4.744577e-07 elapsed=32:13:27 avg_step=1.34s recent_step=1.21s eta=04:56:05 +2026-08-26 12:09:12 | INFO | train rank=0 | step=86730/100000 progress=86.7% loss=0.040657 lr=4.737549e-07 elapsed=32:13:42 avg_step=1.34s recent_step=1.50s eta=04:55:51 +2026-08-26 12:09:27 | INFO | train rank=0 | step=86740/100000 progress=86.7% loss=0.056091 lr=4.730526e-07 elapsed=32:13:57 avg_step=1.34s recent_step=1.45s eta=04:55:38 +2026-08-26 12:09:40 | INFO | train rank=0 | step=86750/100000 progress=86.8% loss=0.043900 lr=4.723509e-07 elapsed=32:14:09 avg_step=1.34s recent_step=1.27s eta=04:55:25 +2026-08-26 12:09:53 | INFO | train rank=0 | step=86760/100000 progress=86.8% loss=0.040939 lr=4.716496e-07 elapsed=32:14:23 avg_step=1.34s recent_step=1.31s eta=04:55:11 +2026-08-26 12:10:08 | INFO | train rank=0 | step=86770/100000 progress=86.8% loss=0.055420 lr=4.709488e-07 elapsed=32:14:38 avg_step=1.34s recent_step=1.54s eta=04:54:58 +2026-08-26 12:10:20 | INFO | train rank=0 | step=86780/100000 progress=86.8% loss=0.049652 lr=4.702485e-07 elapsed=32:14:50 avg_step=1.34s recent_step=1.18s eta=04:54:45 +2026-08-26 12:10:36 | INFO | train rank=0 | step=86790/100000 progress=86.8% loss=0.059402 lr=4.695487e-07 elapsed=32:15:06 avg_step=1.34s recent_step=1.58s eta=04:54:32 +2026-08-26 12:10:50 | INFO | train rank=0 | step=86800/100000 progress=86.8% loss=0.030552 lr=4.688494e-07 elapsed=32:15:20 avg_step=1.34s recent_step=1.47s eta=04:54:18 +2026-08-26 12:11:05 | INFO | train rank=0 | step=86810/100000 progress=86.8% loss=0.027771 lr=4.681505e-07 elapsed=32:15:35 avg_step=1.34s recent_step=1.51s eta=04:54:05 +2026-08-26 12:11:21 | INFO | train rank=0 | step=86820/100000 progress=86.8% loss=0.062531 lr=4.674522e-07 elapsed=32:15:51 avg_step=1.34s recent_step=1.56s eta=04:53:52 +2026-08-26 12:11:33 | INFO | train rank=0 | step=86830/100000 progress=86.8% loss=0.023582 lr=4.667544e-07 elapsed=32:16:03 avg_step=1.34s recent_step=1.20s eta=04:53:39 +2026-08-26 12:11:49 | INFO | train rank=0 | step=86840/100000 progress=86.8% loss=0.042618 lr=4.660571e-07 elapsed=32:16:18 avg_step=1.34s recent_step=1.55s eta=04:53:26 +2026-08-26 12:12:00 | INFO | train rank=0 | step=86850/100000 progress=86.9% loss=0.030899 lr=4.653602e-07 elapsed=32:16:30 avg_step=1.34s recent_step=1.16s eta=04:53:12 +2026-08-26 12:12:15 | INFO | train rank=0 | step=86860/100000 progress=86.9% loss=0.035782 lr=4.646639e-07 elapsed=32:16:45 avg_step=1.34s recent_step=1.54s eta=04:52:59 +2026-08-26 12:12:30 | INFO | train rank=0 | step=86870/100000 progress=86.9% loss=0.039047 lr=4.639681e-07 elapsed=32:17:00 avg_step=1.34s recent_step=1.43s eta=04:52:46 +2026-08-26 12:12:41 | INFO | train rank=0 | step=86880/100000 progress=86.9% loss=0.047989 lr=4.632727e-07 elapsed=32:17:11 avg_step=1.34s recent_step=1.17s eta=04:52:32 +2026-08-26 12:12:53 | INFO | train rank=0 | step=86890/100000 progress=86.9% loss=0.059402 lr=4.625779e-07 elapsed=32:17:23 avg_step=1.34s recent_step=1.13s eta=04:52:18 +2026-08-26 12:13:05 | INFO | train rank=0 | step=86900/100000 progress=86.9% loss=0.035263 lr=4.618835e-07 elapsed=32:17:35 avg_step=1.34s recent_step=1.24s eta=04:52:05 +2026-08-26 12:13:18 | INFO | train rank=0 | step=86910/100000 progress=86.9% loss=0.054707 lr=4.611897e-07 elapsed=32:17:48 avg_step=1.34s recent_step=1.30s eta=04:51:51 +2026-08-26 12:13:33 | INFO | train rank=0 | step=86920/100000 progress=86.9% loss=0.051926 lr=4.604963e-07 elapsed=32:18:03 avg_step=1.34s recent_step=1.53s eta=04:51:38 +2026-08-26 12:13:45 | INFO | train rank=0 | step=86930/100000 progress=86.9% loss=0.054642 lr=4.598035e-07 elapsed=32:18:15 avg_step=1.34s recent_step=1.19s eta=04:51:25 +2026-08-26 12:14:01 | INFO | train rank=0 | step=86940/100000 progress=86.9% loss=0.052795 lr=4.591111e-07 elapsed=32:18:30 avg_step=1.34s recent_step=1.53s eta=04:51:12 +2026-08-26 12:14:15 | INFO | train rank=0 | step=86950/100000 progress=87.0% loss=0.063232 lr=4.584192e-07 elapsed=32:18:45 avg_step=1.34s recent_step=1.42s eta=04:50:58 +2026-08-26 12:14:30 | INFO | train rank=0 | step=86960/100000 progress=87.0% loss=0.047516 lr=4.577278e-07 elapsed=32:19:00 avg_step=1.34s recent_step=1.55s eta=04:50:45 +2026-08-26 12:14:42 | INFO | train rank=0 | step=86970/100000 progress=87.0% loss=0.052887 lr=4.570370e-07 elapsed=32:19:12 avg_step=1.34s recent_step=1.14s eta=04:50:32 +2026-08-26 12:14:56 | INFO | train rank=0 | step=86980/100000 progress=87.0% loss=0.051750 lr=4.563466e-07 elapsed=32:19:25 avg_step=1.34s recent_step=1.39s eta=04:50:18 +2026-08-26 12:15:13 | INFO | train rank=0 | step=86990/100000 progress=87.0% loss=0.040085 lr=4.556567e-07 elapsed=32:19:43 avg_step=1.34s recent_step=1.72s eta=04:50:05 +2026-08-26 12:15:24 | INFO | train rank=0 | step=87000/100000 progress=87.0% loss=0.054260 lr=4.549673e-07 elapsed=32:19:54 avg_step=1.34s recent_step=1.11s eta=04:49:52 +2026-08-26 12:15:36 | INFO | train rank=0 | step=87010/100000 progress=87.0% loss=0.043846 lr=4.542784e-07 elapsed=32:20:06 avg_step=1.34s recent_step=1.21s eta=04:49:38 +2026-08-26 12:15:48 | INFO | train rank=0 | step=87020/100000 progress=87.0% loss=0.054123 lr=4.535900e-07 elapsed=32:20:18 avg_step=1.34s recent_step=1.23s eta=04:49:25 +2026-08-26 12:16:03 | INFO | train rank=0 | step=87030/100000 progress=87.0% loss=0.029297 lr=4.529022e-07 elapsed=32:20:33 avg_step=1.34s recent_step=1.51s eta=04:49:12 +2026-08-26 12:16:16 | INFO | train rank=0 | step=87040/100000 progress=87.0% loss=0.037987 lr=4.522148e-07 elapsed=32:20:46 avg_step=1.34s recent_step=1.25s eta=04:48:58 +2026-08-26 12:16:29 | INFO | train rank=0 | step=87050/100000 progress=87.1% loss=0.086182 lr=4.515279e-07 elapsed=32:20:58 avg_step=1.34s recent_step=1.26s eta=04:48:45 +2026-08-26 12:16:40 | INFO | train rank=0 | step=87060/100000 progress=87.1% loss=0.034836 lr=4.508415e-07 elapsed=32:21:10 avg_step=1.34s recent_step=1.14s eta=04:48:31 +2026-08-26 12:16:52 | INFO | train rank=0 | step=87070/100000 progress=87.1% loss=0.033951 lr=4.501556e-07 elapsed=32:21:22 avg_step=1.34s recent_step=1.17s eta=04:48:17 +2026-08-26 12:17:05 | INFO | train rank=0 | step=87080/100000 progress=87.1% loss=0.048676 lr=4.494701e-07 elapsed=32:21:34 avg_step=1.34s recent_step=1.29s eta=04:48:04 +2026-08-26 12:17:19 | INFO | train rank=0 | step=87090/100000 progress=87.1% loss=0.048477 lr=4.487852e-07 elapsed=32:21:49 avg_step=1.34s recent_step=1.42s eta=04:47:51 +2026-08-26 12:17:33 | INFO | train rank=0 | step=87100/100000 progress=87.1% loss=0.042847 lr=4.481008e-07 elapsed=32:22:03 avg_step=1.34s recent_step=1.39s eta=04:47:37 +2026-08-26 12:17:44 | INFO | train rank=0 | step=87110/100000 progress=87.1% loss=0.027664 lr=4.474169e-07 elapsed=32:22:14 avg_step=1.34s recent_step=1.13s eta=04:47:24 +2026-08-26 12:18:03 | INFO | train rank=0 | step=87120/100000 progress=87.1% loss=0.046417 lr=4.467335e-07 elapsed=32:22:33 avg_step=1.34s recent_step=1.87s eta=04:47:11 +2026-08-26 12:18:14 | INFO | train rank=0 | step=87130/100000 progress=87.1% loss=0.042068 lr=4.460506e-07 elapsed=32:22:44 avg_step=1.34s recent_step=1.12s eta=04:46:57 +2026-08-26 12:18:29 | INFO | train rank=0 | step=87140/100000 progress=87.1% loss=0.028137 lr=4.453682e-07 elapsed=32:22:59 avg_step=1.34s recent_step=1.51s eta=04:46:44 +2026-08-26 12:18:41 | INFO | train rank=0 | step=87150/100000 progress=87.2% loss=0.033920 lr=4.446862e-07 elapsed=32:23:11 avg_step=1.34s recent_step=1.21s eta=04:46:31 +2026-08-26 12:18:58 | INFO | train rank=0 | step=87160/100000 progress=87.2% loss=0.051994 lr=4.440048e-07 elapsed=32:23:28 avg_step=1.34s recent_step=1.70s eta=04:46:18 +2026-08-26 12:19:10 | INFO | train rank=0 | step=87170/100000 progress=87.2% loss=0.056824 lr=4.433239e-07 elapsed=32:23:40 avg_step=1.34s recent_step=1.17s eta=04:46:04 +2026-08-26 12:19:22 | INFO | train rank=0 | step=87180/100000 progress=87.2% loss=0.054550 lr=4.426435e-07 elapsed=32:23:51 avg_step=1.34s recent_step=1.17s eta=04:45:50 +2026-08-26 12:19:38 | INFO | train rank=0 | step=87190/100000 progress=87.2% loss=0.084747 lr=4.419635e-07 elapsed=32:24:08 avg_step=1.34s recent_step=1.64s eta=04:45:38 +2026-08-26 12:19:49 | INFO | train rank=0 | step=87200/100000 progress=87.2% loss=0.055603 lr=4.412841e-07 elapsed=32:24:19 avg_step=1.34s recent_step=1.15s eta=04:45:24 +2026-08-26 12:20:07 | INFO | train rank=0 | step=87210/100000 progress=87.2% loss=0.045166 lr=4.406052e-07 elapsed=32:24:37 avg_step=1.34s recent_step=1.72s eta=04:45:11 +2026-08-26 12:20:19 | INFO | train rank=0 | step=87220/100000 progress=87.2% loss=0.054062 lr=4.399267e-07 elapsed=32:24:49 avg_step=1.34s recent_step=1.28s eta=04:44:58 +2026-08-26 12:20:31 | INFO | train rank=0 | step=87230/100000 progress=87.2% loss=0.041260 lr=4.392488e-07 elapsed=32:25:01 avg_step=1.34s recent_step=1.16s eta=04:44:44 +2026-08-26 12:20:43 | INFO | train rank=0 | step=87240/100000 progress=87.2% loss=0.034958 lr=4.385714e-07 elapsed=32:25:13 avg_step=1.34s recent_step=1.22s eta=04:44:30 +2026-08-26 12:21:02 | INFO | train rank=0 | step=87250/100000 progress=87.2% loss=0.050629 lr=4.378944e-07 elapsed=32:25:32 avg_step=1.34s recent_step=1.90s eta=04:44:18 +2026-08-26 12:21:15 | INFO | train rank=0 | step=87260/100000 progress=87.3% loss=0.027565 lr=4.372180e-07 elapsed=32:25:45 avg_step=1.34s recent_step=1.28s eta=04:44:04 +2026-08-26 12:21:27 | INFO | train rank=0 | step=87270/100000 progress=87.3% loss=0.041580 lr=4.365421e-07 elapsed=32:25:56 avg_step=1.34s recent_step=1.16s eta=04:43:51 +2026-08-26 12:21:39 | INFO | train rank=0 | step=87280/100000 progress=87.3% loss=0.079437 lr=4.358666e-07 elapsed=32:26:09 avg_step=1.34s recent_step=1.25s eta=04:43:37 +2026-08-26 12:21:51 | INFO | train rank=0 | step=87290/100000 progress=87.3% loss=0.046539 lr=4.351917e-07 elapsed=32:26:21 avg_step=1.34s recent_step=1.18s eta=04:43:24 +2026-08-26 12:22:11 | INFO | train rank=0 | step=87300/100000 progress=87.3% loss=0.049316 lr=4.345172e-07 elapsed=32:26:41 avg_step=1.34s recent_step=2.00s eta=04:43:11 +2026-08-26 12:22:22 | INFO | train rank=0 | step=87310/100000 progress=87.3% loss=0.037827 lr=4.338433e-07 elapsed=32:26:52 avg_step=1.34s recent_step=1.15s eta=04:42:58 +2026-08-26 12:22:34 | INFO | train rank=0 | step=87320/100000 progress=87.3% loss=0.047958 lr=4.331699e-07 elapsed=32:27:04 avg_step=1.34s recent_step=1.17s eta=04:42:44 +2026-08-26 12:22:47 | INFO | train rank=0 | step=87330/100000 progress=87.3% loss=0.048080 lr=4.324969e-07 elapsed=32:27:17 avg_step=1.34s recent_step=1.29s eta=04:42:30 +2026-08-26 12:23:02 | INFO | train rank=0 | step=87340/100000 progress=87.3% loss=0.042618 lr=4.318245e-07 elapsed=32:27:32 avg_step=1.34s recent_step=1.47s eta=04:42:17 +2026-08-26 12:23:13 | INFO | train rank=0 | step=87350/100000 progress=87.4% loss=0.051437 lr=4.311525e-07 elapsed=32:27:43 avg_step=1.34s recent_step=1.15s eta=04:42:04 +2026-08-26 12:23:29 | INFO | train rank=0 | step=87360/100000 progress=87.4% loss=0.043304 lr=4.304811e-07 elapsed=32:27:59 avg_step=1.34s recent_step=1.58s eta=04:41:51 +2026-08-26 12:23:41 | INFO | train rank=0 | step=87370/100000 progress=87.4% loss=0.055298 lr=4.298101e-07 elapsed=32:28:11 avg_step=1.34s recent_step=1.19s eta=04:41:37 +2026-08-26 12:23:53 | INFO | train rank=0 | step=87380/100000 progress=87.4% loss=0.048981 lr=4.291397e-07 elapsed=32:28:23 avg_step=1.34s recent_step=1.20s eta=04:41:23 +2026-08-26 12:24:05 | INFO | train rank=0 | step=87390/100000 progress=87.4% loss=0.029099 lr=4.284697e-07 elapsed=32:28:35 avg_step=1.34s recent_step=1.19s eta=04:41:10 +2026-08-26 12:24:19 | INFO | train rank=0 | step=87400/100000 progress=87.4% loss=0.032791 lr=4.278003e-07 elapsed=32:28:49 avg_step=1.34s recent_step=1.45s eta=04:40:57 +2026-08-26 12:24:32 | INFO | train rank=0 | step=87410/100000 progress=87.4% loss=0.071320 lr=4.271314e-07 elapsed=32:29:02 avg_step=1.34s recent_step=1.25s eta=04:40:43 +2026-08-26 12:24:52 | INFO | train rank=0 | step=87420/100000 progress=87.4% loss=0.042496 lr=4.264629e-07 elapsed=32:29:22 avg_step=1.34s recent_step=2.00s eta=04:40:31 +2026-08-26 12:25:04 | INFO | train rank=0 | step=87430/100000 progress=87.4% loss=0.043442 lr=4.257950e-07 elapsed=32:29:33 avg_step=1.34s recent_step=1.18s eta=04:40:17 +2026-08-26 12:25:15 | INFO | train rank=0 | step=87440/100000 progress=87.4% loss=0.022675 lr=4.251275e-07 elapsed=32:29:45 avg_step=1.34s recent_step=1.13s eta=04:40:03 +2026-08-26 12:25:27 | INFO | train rank=0 | step=87450/100000 progress=87.5% loss=0.048157 lr=4.244606e-07 elapsed=32:29:57 avg_step=1.34s recent_step=1.17s eta=04:39:50 +2026-08-26 12:25:45 | INFO | train rank=0 | step=87460/100000 progress=87.5% loss=0.060226 lr=4.237941e-07 elapsed=32:30:15 avg_step=1.34s recent_step=1.85s eta=04:39:37 +2026-08-26 12:25:57 | INFO | train rank=0 | step=87470/100000 progress=87.5% loss=0.076416 lr=4.231282e-07 elapsed=32:30:27 avg_step=1.34s recent_step=1.21s eta=04:39:24 +2026-08-26 12:26:10 | INFO | train rank=0 | step=87480/100000 progress=87.5% loss=0.049400 lr=4.224628e-07 elapsed=32:30:40 avg_step=1.34s recent_step=1.29s eta=04:39:10 +2026-08-26 12:26:22 | INFO | train rank=0 | step=87490/100000 progress=87.5% loss=0.056366 lr=4.217978e-07 elapsed=32:30:52 avg_step=1.34s recent_step=1.22s eta=04:38:57 +2026-08-26 12:26:34 | INFO | train rank=0 | step=87500/100000 progress=87.5% loss=0.045197 lr=4.211334e-07 elapsed=32:31:04 avg_step=1.34s recent_step=1.16s eta=04:38:43 +2026-08-26 12:26:48 | INFO | train rank=0 | step=87510/100000 progress=87.5% loss=0.069519 lr=4.204694e-07 elapsed=32:31:18 avg_step=1.34s recent_step=1.44s eta=04:38:30 +2026-08-26 12:27:01 | INFO | train rank=0 | step=87520/100000 progress=87.5% loss=0.033432 lr=4.198060e-07 elapsed=32:31:31 avg_step=1.34s recent_step=1.26s eta=04:38:16 +2026-08-26 12:27:14 | INFO | train rank=0 | step=87530/100000 progress=87.5% loss=0.047668 lr=4.191430e-07 elapsed=32:31:44 avg_step=1.34s recent_step=1.33s eta=04:38:03 +2026-08-26 12:27:29 | INFO | train rank=0 | step=87540/100000 progress=87.5% loss=0.035828 lr=4.184806e-07 elapsed=32:31:59 avg_step=1.34s recent_step=1.48s eta=04:37:50 +2026-08-26 12:27:44 | INFO | train rank=0 | step=87550/100000 progress=87.5% loss=0.072296 lr=4.178187e-07 elapsed=32:32:14 avg_step=1.34s recent_step=1.47s eta=04:37:37 +2026-08-26 12:27:55 | INFO | train rank=0 | step=87560/100000 progress=87.6% loss=0.076904 lr=4.171572e-07 elapsed=32:32:25 avg_step=1.34s recent_step=1.14s eta=04:37:23 +2026-08-26 12:28:07 | INFO | train rank=0 | step=87570/100000 progress=87.6% loss=0.037903 lr=4.164963e-07 elapsed=32:32:37 avg_step=1.34s recent_step=1.15s eta=04:37:09 +2026-08-26 12:28:24 | INFO | train rank=0 | step=87580/100000 progress=87.6% loss=0.042267 lr=4.158359e-07 elapsed=32:32:54 avg_step=1.34s recent_step=1.73s eta=04:36:56 +2026-08-26 12:28:39 | INFO | train rank=0 | step=87590/100000 progress=87.6% loss=0.052399 lr=4.151759e-07 elapsed=32:33:09 avg_step=1.34s recent_step=1.52s eta=04:36:43 +2026-08-26 12:28:51 | INFO | train rank=0 | step=87600/100000 progress=87.6% loss=0.027557 lr=4.145165e-07 elapsed=32:33:21 avg_step=1.34s recent_step=1.18s eta=04:36:30 +2026-08-26 12:29:04 | INFO | train rank=0 | step=87610/100000 progress=87.6% loss=0.058456 lr=4.138576e-07 elapsed=32:33:34 avg_step=1.34s recent_step=1.28s eta=04:36:16 +2026-08-26 12:29:19 | INFO | train rank=0 | step=87620/100000 progress=87.6% loss=0.047440 lr=4.131991e-07 elapsed=32:33:49 avg_step=1.34s recent_step=1.55s eta=04:36:03 +2026-08-26 12:29:31 | INFO | train rank=0 | step=87630/100000 progress=87.6% loss=0.064178 lr=4.125412e-07 elapsed=32:34:01 avg_step=1.34s recent_step=1.20s eta=04:35:50 +2026-08-26 12:29:44 | INFO | train rank=0 | step=87640/100000 progress=87.6% loss=0.043770 lr=4.118838e-07 elapsed=32:34:14 avg_step=1.34s recent_step=1.31s eta=04:35:36 +2026-08-26 12:30:02 | INFO | train rank=0 | step=87650/100000 progress=87.6% loss=0.042038 lr=4.112269e-07 elapsed=32:34:32 avg_step=1.34s recent_step=1.80s eta=04:35:23 +2026-08-26 12:30:16 | INFO | train rank=0 | step=87660/100000 progress=87.7% loss=0.043961 lr=4.105705e-07 elapsed=32:34:46 avg_step=1.34s recent_step=1.40s eta=04:35:10 +2026-08-26 12:30:27 | INFO | train rank=0 | step=87670/100000 progress=87.7% loss=0.035721 lr=4.099145e-07 elapsed=32:34:57 avg_step=1.34s recent_step=1.10s eta=04:34:56 +2026-08-26 12:30:42 | INFO | train rank=0 | step=87680/100000 progress=87.7% loss=0.050232 lr=4.092591e-07 elapsed=32:35:12 avg_step=1.34s recent_step=1.44s eta=04:34:43 +2026-08-26 12:30:55 | INFO | train rank=0 | step=87690/100000 progress=87.7% loss=0.044586 lr=4.086042e-07 elapsed=32:35:25 avg_step=1.34s recent_step=1.34s eta=04:34:30 +2026-08-26 12:31:07 | INFO | train rank=0 | step=87700/100000 progress=87.7% loss=0.048279 lr=4.079498e-07 elapsed=32:35:37 avg_step=1.34s recent_step=1.23s eta=04:34:16 +2026-08-26 12:31:22 | INFO | train rank=0 | step=87710/100000 progress=87.7% loss=0.031052 lr=4.072959e-07 elapsed=32:35:52 avg_step=1.34s recent_step=1.44s eta=04:34:03 +2026-08-26 12:31:34 | INFO | train rank=0 | step=87720/100000 progress=87.7% loss=0.063141 lr=4.066425e-07 elapsed=32:36:03 avg_step=1.34s recent_step=1.18s eta=04:33:49 +2026-08-26 12:31:47 | INFO | train rank=0 | step=87730/100000 progress=87.7% loss=0.033325 lr=4.059896e-07 elapsed=32:36:16 avg_step=1.34s recent_step=1.30s eta=04:33:36 +2026-08-26 12:31:58 | INFO | train rank=0 | step=87740/100000 progress=87.7% loss=0.073029 lr=4.053372e-07 elapsed=32:36:28 avg_step=1.34s recent_step=1.17s eta=04:33:22 +2026-08-26 12:32:11 | INFO | train rank=0 | step=87750/100000 progress=87.8% loss=0.056412 lr=4.046853e-07 elapsed=32:36:41 avg_step=1.34s recent_step=1.28s eta=04:33:09 +2026-08-26 12:32:26 | INFO | train rank=0 | step=87760/100000 progress=87.8% loss=0.030708 lr=4.040339e-07 elapsed=32:36:56 avg_step=1.34s recent_step=1.47s eta=04:32:56 +2026-08-26 12:32:39 | INFO | train rank=0 | step=87770/100000 progress=87.8% loss=0.027390 lr=4.033830e-07 elapsed=32:37:09 avg_step=1.34s recent_step=1.29s eta=04:32:42 +2026-08-26 12:32:56 | INFO | train rank=0 | step=87780/100000 progress=87.8% loss=0.075134 lr=4.027326e-07 elapsed=32:37:26 avg_step=1.34s recent_step=1.75s eta=04:32:29 +2026-08-26 12:33:08 | INFO | train rank=0 | step=87790/100000 progress=87.8% loss=0.032539 lr=4.020827e-07 elapsed=32:37:38 avg_step=1.34s recent_step=1.15s eta=04:32:16 +2026-08-26 12:33:19 | INFO | train rank=0 | step=87800/100000 progress=87.8% loss=0.078644 lr=4.014333e-07 elapsed=32:37:49 avg_step=1.34s recent_step=1.16s eta=04:32:02 +2026-08-26 12:33:32 | INFO | train rank=0 | step=87810/100000 progress=87.8% loss=0.045151 lr=4.007844e-07 elapsed=32:38:02 avg_step=1.34s recent_step=1.27s eta=04:31:49 +2026-08-26 12:33:44 | INFO | train rank=0 | step=87820/100000 progress=87.8% loss=0.070648 lr=4.001360e-07 elapsed=32:38:14 avg_step=1.34s recent_step=1.19s eta=04:31:35 +2026-08-26 12:33:57 | INFO | train rank=0 | step=87830/100000 progress=87.8% loss=0.039642 lr=3.994881e-07 elapsed=32:38:27 avg_step=1.34s recent_step=1.30s eta=04:31:22 +2026-08-26 12:34:09 | INFO | train rank=0 | step=87840/100000 progress=87.8% loss=0.041016 lr=3.988408e-07 elapsed=32:38:39 avg_step=1.34s recent_step=1.15s eta=04:31:08 +2026-08-26 12:34:21 | INFO | train rank=0 | step=87850/100000 progress=87.8% loss=0.050934 lr=3.981939e-07 elapsed=32:38:51 avg_step=1.34s recent_step=1.24s eta=04:30:55 +2026-08-26 12:34:33 | INFO | train rank=0 | step=87860/100000 progress=87.9% loss=0.044739 lr=3.975475e-07 elapsed=32:39:03 avg_step=1.34s recent_step=1.18s eta=04:30:41 +2026-08-26 12:34:45 | INFO | train rank=0 | step=87870/100000 progress=87.9% loss=0.025288 lr=3.969017e-07 elapsed=32:39:14 avg_step=1.34s recent_step=1.17s eta=04:30:27 +2026-08-26 12:34:58 | INFO | train rank=0 | step=87880/100000 progress=87.9% loss=0.040573 lr=3.962563e-07 elapsed=32:39:28 avg_step=1.34s recent_step=1.36s eta=04:30:14 +2026-08-26 12:35:12 | INFO | train rank=0 | step=87890/100000 progress=87.9% loss=0.080902 lr=3.956114e-07 elapsed=32:39:41 avg_step=1.34s recent_step=1.34s eta=04:30:01 +2026-08-26 12:35:25 | INFO | train rank=0 | step=87900/100000 progress=87.9% loss=0.078430 lr=3.949671e-07 elapsed=32:39:55 avg_step=1.34s recent_step=1.35s eta=04:29:47 +2026-08-26 12:35:37 | INFO | train rank=0 | step=87910/100000 progress=87.9% loss=0.056702 lr=3.943232e-07 elapsed=32:40:07 avg_step=1.34s recent_step=1.24s eta=04:29:34 +2026-08-26 12:35:51 | INFO | train rank=0 | step=87920/100000 progress=87.9% loss=0.029945 lr=3.936799e-07 elapsed=32:40:21 avg_step=1.34s recent_step=1.37s eta=04:29:20 +2026-08-26 12:36:05 | INFO | train rank=0 | step=87930/100000 progress=87.9% loss=0.047867 lr=3.930370e-07 elapsed=32:40:35 avg_step=1.34s recent_step=1.42s eta=04:29:07 +2026-08-26 12:36:19 | INFO | train rank=0 | step=87940/100000 progress=87.9% loss=0.045517 lr=3.923947e-07 elapsed=32:40:49 avg_step=1.34s recent_step=1.39s eta=04:28:54 +2026-08-26 12:36:34 | INFO | train rank=0 | step=87950/100000 progress=87.9% loss=0.053604 lr=3.917529e-07 elapsed=32:41:04 avg_step=1.34s recent_step=1.52s eta=04:28:41 +2026-08-26 12:36:46 | INFO | train rank=0 | step=87960/100000 progress=88.0% loss=0.040970 lr=3.911115e-07 elapsed=32:41:16 avg_step=1.34s recent_step=1.16s eta=04:28:27 +2026-08-26 12:37:02 | INFO | train rank=0 | step=87970/100000 progress=88.0% loss=0.054886 lr=3.904707e-07 elapsed=32:41:32 avg_step=1.34s recent_step=1.59s eta=04:28:14 +2026-08-26 12:37:15 | INFO | train rank=0 | step=87980/100000 progress=88.0% loss=0.064606 lr=3.898304e-07 elapsed=32:41:44 avg_step=1.34s recent_step=1.27s eta=04:28:01 +2026-08-26 12:37:28 | INFO | train rank=0 | step=87990/100000 progress=88.0% loss=0.055481 lr=3.891905e-07 elapsed=32:41:58 avg_step=1.34s recent_step=1.31s eta=04:27:47 +2026-08-26 12:37:44 | INFO | train rank=0 | step=88000/100000 progress=88.0% loss=0.028519 lr=3.885512e-07 elapsed=32:42:14 avg_step=1.34s recent_step=1.62s eta=04:27:34 +2026-08-26 12:37:57 | INFO | train rank=0 | step=88010/100000 progress=88.0% loss=0.042618 lr=3.879124e-07 elapsed=32:42:27 avg_step=1.34s recent_step=1.30s eta=04:27:21 +2026-08-26 12:38:11 | INFO | train rank=0 | step=88020/100000 progress=88.0% loss=0.056396 lr=3.872741e-07 elapsed=32:42:40 avg_step=1.34s recent_step=1.36s eta=04:27:07 +2026-08-26 12:38:24 | INFO | train rank=0 | step=88030/100000 progress=88.0% loss=0.035492 lr=3.866363e-07 elapsed=32:42:54 avg_step=1.34s recent_step=1.36s eta=04:26:54 +2026-08-26 12:38:37 | INFO | train rank=0 | step=88040/100000 progress=88.0% loss=0.033417 lr=3.859990e-07 elapsed=32:43:07 avg_step=1.34s recent_step=1.28s eta=04:26:41 +2026-08-26 12:38:51 | INFO | train rank=0 | step=88050/100000 progress=88.0% loss=0.037182 lr=3.853622e-07 elapsed=32:43:21 avg_step=1.34s recent_step=1.40s eta=04:26:27 +2026-08-26 12:39:03 | INFO | train rank=0 | step=88060/100000 progress=88.1% loss=0.054642 lr=3.847259e-07 elapsed=32:43:33 avg_step=1.34s recent_step=1.25s eta=04:26:14 +2026-08-26 12:39:18 | INFO | train rank=0 | step=88070/100000 progress=88.1% loss=0.035233 lr=3.840901e-07 elapsed=32:43:47 avg_step=1.34s recent_step=1.41s eta=04:26:01 +2026-08-26 12:39:33 | INFO | train rank=0 | step=88080/100000 progress=88.1% loss=0.042816 lr=3.834549e-07 elapsed=32:44:03 avg_step=1.34s recent_step=1.51s eta=04:25:47 +2026-08-26 12:39:47 | INFO | train rank=0 | step=88090/100000 progress=88.1% loss=0.080505 lr=3.828201e-07 elapsed=32:44:17 avg_step=1.34s recent_step=1.46s eta=04:25:34 +2026-08-26 12:40:02 | INFO | train rank=0 | step=88100/100000 progress=88.1% loss=0.060043 lr=3.821858e-07 elapsed=32:44:32 avg_step=1.34s recent_step=1.47s eta=04:25:21 +2026-08-26 12:40:14 | INFO | train rank=0 | step=88110/100000 progress=88.1% loss=0.050797 lr=3.815520e-07 elapsed=32:44:44 avg_step=1.34s recent_step=1.23s eta=04:25:07 +2026-08-26 12:40:29 | INFO | train rank=0 | step=88120/100000 progress=88.1% loss=0.062866 lr=3.809188e-07 elapsed=32:44:59 avg_step=1.34s recent_step=1.45s eta=04:24:54 +2026-08-26 12:40:41 | INFO | train rank=0 | step=88130/100000 progress=88.1% loss=0.039932 lr=3.802860e-07 elapsed=32:45:11 avg_step=1.34s recent_step=1.24s eta=04:24:41 +2026-08-26 12:40:54 | INFO | train rank=0 | step=88140/100000 progress=88.1% loss=0.045967 lr=3.796538e-07 elapsed=32:45:24 avg_step=1.34s recent_step=1.27s eta=04:24:27 +2026-08-26 12:41:07 | INFO | train rank=0 | step=88150/100000 progress=88.1% loss=0.061523 lr=3.790220e-07 elapsed=32:45:36 avg_step=1.34s recent_step=1.27s eta=04:24:14 +2026-08-26 12:41:20 | INFO | train rank=0 | step=88160/100000 progress=88.2% loss=0.057419 lr=3.783908e-07 elapsed=32:45:50 avg_step=1.34s recent_step=1.35s eta=04:24:00 +2026-08-26 12:41:33 | INFO | train rank=0 | step=88170/100000 progress=88.2% loss=0.052078 lr=3.777601e-07 elapsed=32:46:03 avg_step=1.34s recent_step=1.30s eta=04:23:47 +2026-08-26 12:41:46 | INFO | train rank=0 | step=88180/100000 progress=88.2% loss=0.061661 lr=3.771298e-07 elapsed=32:46:16 avg_step=1.34s recent_step=1.27s eta=04:23:33 +2026-08-26 12:41:58 | INFO | train rank=0 | step=88190/100000 progress=88.2% loss=0.031746 lr=3.765001e-07 elapsed=32:46:28 avg_step=1.34s recent_step=1.26s eta=04:23:20 +2026-08-26 12:42:12 | INFO | train rank=0 | step=88200/100000 progress=88.2% loss=0.035736 lr=3.758709e-07 elapsed=32:46:42 avg_step=1.34s recent_step=1.41s eta=04:23:07 +2026-08-26 12:42:25 | INFO | train rank=0 | step=88210/100000 progress=88.2% loss=0.042114 lr=3.752422e-07 elapsed=32:46:55 avg_step=1.34s recent_step=1.25s eta=04:22:53 +2026-08-26 12:42:38 | INFO | train rank=0 | step=88220/100000 progress=88.2% loss=0.084381 lr=3.746140e-07 elapsed=32:47:08 avg_step=1.34s recent_step=1.28s eta=04:22:40 +2026-08-26 12:42:50 | INFO | train rank=0 | step=88230/100000 progress=88.2% loss=0.065887 lr=3.739863e-07 elapsed=32:47:20 avg_step=1.34s recent_step=1.25s eta=04:22:26 +2026-08-26 12:43:03 | INFO | train rank=0 | step=88240/100000 progress=88.2% loss=0.054459 lr=3.733591e-07 elapsed=32:47:33 avg_step=1.34s recent_step=1.27s eta=04:22:13 +2026-08-26 12:43:19 | INFO | train rank=0 | step=88250/100000 progress=88.2% loss=0.050522 lr=3.727324e-07 elapsed=32:47:49 avg_step=1.34s recent_step=1.60s eta=04:22:00 +2026-08-26 12:43:31 | INFO | train rank=0 | step=88260/100000 progress=88.3% loss=0.038643 lr=3.721062e-07 elapsed=32:48:01 avg_step=1.34s recent_step=1.20s eta=04:21:46 +2026-08-26 12:43:46 | INFO | train rank=0 | step=88270/100000 progress=88.3% loss=0.050507 lr=3.714805e-07 elapsed=32:48:16 avg_step=1.34s recent_step=1.53s eta=04:21:33 +2026-08-26 12:44:00 | INFO | train rank=0 | step=88280/100000 progress=88.3% loss=0.038681 lr=3.708554e-07 elapsed=32:48:30 avg_step=1.34s recent_step=1.40s eta=04:21:20 +2026-08-26 12:44:14 | INFO | train rank=0 | step=88290/100000 progress=88.3% loss=0.057648 lr=3.702307e-07 elapsed=32:48:44 avg_step=1.34s recent_step=1.42s eta=04:21:07 +2026-08-26 12:44:26 | INFO | train rank=0 | step=88300/100000 progress=88.3% loss=0.058044 lr=3.696065e-07 elapsed=32:48:56 avg_step=1.34s recent_step=1.18s eta=04:20:53 +2026-08-26 12:44:43 | INFO | train rank=0 | step=88310/100000 progress=88.3% loss=0.038239 lr=3.689829e-07 elapsed=32:49:13 avg_step=1.34s recent_step=1.66s eta=04:20:40 +2026-08-26 12:44:55 | INFO | train rank=0 | step=88320/100000 progress=88.3% loss=0.034592 lr=3.683597e-07 elapsed=32:49:25 avg_step=1.34s recent_step=1.25s eta=04:20:26 +2026-08-26 12:45:13 | INFO | train rank=0 | step=88330/100000 progress=88.3% loss=0.040222 lr=3.677371e-07 elapsed=32:49:43 avg_step=1.34s recent_step=1.82s eta=04:20:14 +2026-08-26 12:45:25 | INFO | train rank=0 | step=88340/100000 progress=88.3% loss=0.044243 lr=3.671150e-07 elapsed=32:49:55 avg_step=1.34s recent_step=1.19s eta=04:20:00 +2026-08-26 12:45:37 | INFO | train rank=0 | step=88350/100000 progress=88.3% loss=0.044106 lr=3.664933e-07 elapsed=32:50:07 avg_step=1.34s recent_step=1.20s eta=04:19:47 +2026-08-26 12:45:49 | INFO | train rank=0 | step=88360/100000 progress=88.4% loss=0.077454 lr=3.658722e-07 elapsed=32:50:19 avg_step=1.34s recent_step=1.23s eta=04:19:33 +2026-08-26 12:46:01 | INFO | train rank=0 | step=88370/100000 progress=88.4% loss=0.030167 lr=3.652516e-07 elapsed=32:50:31 avg_step=1.34s recent_step=1.14s eta=04:19:19 +2026-08-26 12:46:13 | INFO | train rank=0 | step=88380/100000 progress=88.4% loss=0.073242 lr=3.646315e-07 elapsed=32:50:42 avg_step=1.34s recent_step=1.17s eta=04:19:06 +2026-08-26 12:46:25 | INFO | train rank=0 | step=88390/100000 progress=88.4% loss=0.041626 lr=3.640119e-07 elapsed=32:50:55 avg_step=1.34s recent_step=1.29s eta=04:18:52 +2026-08-26 12:46:37 | INFO | train rank=0 | step=88400/100000 progress=88.4% loss=0.045258 lr=3.633928e-07 elapsed=32:51:07 avg_step=1.34s recent_step=1.19s eta=04:18:39 +2026-08-26 12:46:49 | INFO | train rank=0 | step=88410/100000 progress=88.4% loss=0.034958 lr=3.627742e-07 elapsed=32:51:19 avg_step=1.34s recent_step=1.21s eta=04:18:25 +2026-08-26 12:47:03 | INFO | train rank=0 | step=88420/100000 progress=88.4% loss=0.062454 lr=3.621562e-07 elapsed=32:51:33 avg_step=1.34s recent_step=1.35s eta=04:18:12 +2026-08-26 12:47:15 | INFO | train rank=0 | step=88430/100000 progress=88.4% loss=0.051178 lr=3.615386e-07 elapsed=32:51:45 avg_step=1.34s recent_step=1.22s eta=04:17:58 +2026-08-26 12:47:33 | INFO | train rank=0 | step=88440/100000 progress=88.4% loss=0.060333 lr=3.609215e-07 elapsed=32:52:03 avg_step=1.34s recent_step=1.83s eta=04:17:46 +2026-08-26 12:47:46 | INFO | train rank=0 | step=88450/100000 progress=88.4% loss=0.038834 lr=3.603050e-07 elapsed=32:52:16 avg_step=1.34s recent_step=1.29s eta=04:17:32 +2026-08-26 12:48:00 | INFO | train rank=0 | step=88460/100000 progress=88.5% loss=0.040146 lr=3.596889e-07 elapsed=32:52:30 avg_step=1.34s recent_step=1.36s eta=04:17:19 +2026-08-26 12:48:17 | INFO | train rank=0 | step=88470/100000 progress=88.5% loss=0.059082 lr=3.590734e-07 elapsed=32:52:47 avg_step=1.34s recent_step=1.70s eta=04:17:06 +2026-08-26 12:48:29 | INFO | train rank=0 | step=88480/100000 progress=88.5% loss=0.043060 lr=3.584583e-07 elapsed=32:52:59 avg_step=1.34s recent_step=1.17s eta=04:16:52 +2026-08-26 12:48:41 | INFO | train rank=0 | step=88490/100000 progress=88.5% loss=0.029434 lr=3.578438e-07 elapsed=32:53:11 avg_step=1.34s recent_step=1.24s eta=04:16:39 +2026-08-26 12:48:54 | INFO | train rank=0 | step=88500/100000 progress=88.5% loss=0.039185 lr=3.572298e-07 elapsed=32:53:24 avg_step=1.34s recent_step=1.26s eta=04:16:25 +2026-08-26 12:49:06 | INFO | train rank=0 | step=88510/100000 progress=88.5% loss=0.067398 lr=3.566163e-07 elapsed=32:53:35 avg_step=1.34s recent_step=1.19s eta=04:16:12 +2026-08-26 12:49:22 | INFO | train rank=0 | step=88520/100000 progress=88.5% loss=0.036812 lr=3.560033e-07 elapsed=32:53:52 avg_step=1.34s recent_step=1.61s eta=04:15:59 +2026-08-26 12:49:37 | INFO | train rank=0 | step=88530/100000 progress=88.5% loss=0.045448 lr=3.553908e-07 elapsed=32:54:07 avg_step=1.34s recent_step=1.52s eta=04:15:46 +2026-08-26 12:49:49 | INFO | train rank=0 | step=88540/100000 progress=88.5% loss=0.050781 lr=3.547788e-07 elapsed=32:54:19 avg_step=1.34s recent_step=1.18s eta=04:15:32 +2026-08-26 12:50:04 | INFO | train rank=0 | step=88550/100000 progress=88.5% loss=0.035286 lr=3.541673e-07 elapsed=32:54:34 avg_step=1.34s recent_step=1.50s eta=04:15:19 +2026-08-26 12:50:16 | INFO | train rank=0 | step=88560/100000 progress=88.6% loss=0.026947 lr=3.535564e-07 elapsed=32:54:46 avg_step=1.34s recent_step=1.27s eta=04:15:05 +2026-08-26 12:50:28 | INFO | train rank=0 | step=88570/100000 progress=88.6% loss=0.052505 lr=3.529459e-07 elapsed=32:54:58 avg_step=1.34s recent_step=1.21s eta=04:14:52 +2026-08-26 12:50:40 | INFO | train rank=0 | step=88580/100000 progress=88.6% loss=0.044174 lr=3.523360e-07 elapsed=32:55:10 avg_step=1.34s recent_step=1.15s eta=04:14:38 +2026-08-26 12:50:54 | INFO | train rank=0 | step=88590/100000 progress=88.6% loss=0.048706 lr=3.517265e-07 elapsed=32:55:24 avg_step=1.34s recent_step=1.44s eta=04:14:25 +2026-08-26 12:51:13 | INFO | train rank=0 | step=88600/100000 progress=88.6% loss=0.030075 lr=3.511176e-07 elapsed=32:55:43 avg_step=1.34s recent_step=1.91s eta=04:14:12 +2026-08-26 12:51:25 | INFO | train rank=0 | step=88610/100000 progress=88.6% loss=0.041580 lr=3.505091e-07 elapsed=32:55:55 avg_step=1.34s recent_step=1.17s eta=04:13:59 +2026-08-26 12:51:37 | INFO | train rank=0 | step=88620/100000 progress=88.6% loss=0.046661 lr=3.499012e-07 elapsed=32:56:07 avg_step=1.34s recent_step=1.18s eta=04:13:45 +2026-08-26 12:51:54 | INFO | train rank=0 | step=88630/100000 progress=88.6% loss=0.061523 lr=3.492938e-07 elapsed=32:56:24 avg_step=1.34s recent_step=1.69s eta=04:13:32 +2026-08-26 12:52:06 | INFO | train rank=0 | step=88640/100000 progress=88.6% loss=0.039886 lr=3.486869e-07 elapsed=32:56:36 avg_step=1.34s recent_step=1.19s eta=04:13:19 +2026-08-26 12:52:19 | INFO | train rank=0 | step=88650/100000 progress=88.6% loss=0.045235 lr=3.480805e-07 elapsed=32:56:48 avg_step=1.34s recent_step=1.29s eta=04:13:05 +2026-08-26 12:52:31 | INFO | train rank=0 | step=88660/100000 progress=88.7% loss=0.084473 lr=3.474746e-07 elapsed=32:57:00 avg_step=1.34s recent_step=1.19s eta=04:12:52 +2026-08-26 12:52:44 | INFO | train rank=0 | step=88670/100000 progress=88.7% loss=0.027313 lr=3.468693e-07 elapsed=32:57:14 avg_step=1.34s recent_step=1.38s eta=04:12:38 +2026-08-26 12:52:57 | INFO | train rank=0 | step=88680/100000 progress=88.7% loss=0.045914 lr=3.462644e-07 elapsed=32:57:27 avg_step=1.34s recent_step=1.24s eta=04:12:25 +2026-08-26 12:53:09 | INFO | train rank=0 | step=88690/100000 progress=88.7% loss=0.044525 lr=3.456600e-07 elapsed=32:57:39 avg_step=1.34s recent_step=1.24s eta=04:12:11 +2026-08-26 12:53:29 | INFO | train rank=0 | step=88700/100000 progress=88.7% loss=0.071045 lr=3.450562e-07 elapsed=32:57:59 avg_step=1.34s recent_step=1.95s eta=04:11:59 +2026-08-26 12:53:41 | INFO | train rank=0 | step=88710/100000 progress=88.7% loss=0.053726 lr=3.444528e-07 elapsed=32:58:10 avg_step=1.34s recent_step=1.19s eta=04:11:45 +2026-08-26 12:53:52 | INFO | train rank=0 | step=88720/100000 progress=88.7% loss=0.039993 lr=3.438500e-07 elapsed=32:58:22 avg_step=1.34s recent_step=1.18s eta=04:11:32 +2026-08-26 12:54:05 | INFO | train rank=0 | step=88730/100000 progress=88.7% loss=0.055695 lr=3.432477e-07 elapsed=32:58:35 avg_step=1.34s recent_step=1.25s eta=04:11:18 +2026-08-26 12:54:17 | INFO | train rank=0 | step=88740/100000 progress=88.7% loss=0.048981 lr=3.426459e-07 elapsed=32:58:47 avg_step=1.34s recent_step=1.23s eta=04:11:05 +2026-08-26 12:54:32 | INFO | train rank=0 | step=88750/100000 progress=88.8% loss=0.040276 lr=3.420446e-07 elapsed=32:59:02 avg_step=1.34s recent_step=1.49s eta=04:10:51 +2026-08-26 12:54:47 | INFO | train rank=0 | step=88760/100000 progress=88.8% loss=0.033646 lr=3.414438e-07 elapsed=32:59:17 avg_step=1.34s recent_step=1.48s eta=04:10:38 +2026-08-26 12:54:59 | INFO | train rank=0 | step=88770/100000 progress=88.8% loss=0.031128 lr=3.408435e-07 elapsed=32:59:29 avg_step=1.34s recent_step=1.23s eta=04:10:25 +2026-08-26 12:55:14 | INFO | train rank=0 | step=88780/100000 progress=88.8% loss=0.037430 lr=3.402437e-07 elapsed=32:59:44 avg_step=1.34s recent_step=1.46s eta=04:10:11 +2026-08-26 12:55:28 | INFO | train rank=0 | step=88790/100000 progress=88.8% loss=0.054901 lr=3.396444e-07 elapsed=32:59:58 avg_step=1.34s recent_step=1.41s eta=04:09:58 +2026-08-26 12:55:40 | INFO | train rank=0 | step=88800/100000 progress=88.8% loss=0.054855 lr=3.390457e-07 elapsed=33:00:09 avg_step=1.34s recent_step=1.16s eta=04:09:45 +2026-08-26 12:55:56 | INFO | train rank=0 | step=88810/100000 progress=88.8% loss=0.032036 lr=3.384474e-07 elapsed=33:00:26 avg_step=1.34s recent_step=1.67s eta=04:09:32 +2026-08-26 12:56:08 | INFO | train rank=0 | step=88820/100000 progress=88.8% loss=0.040031 lr=3.378497e-07 elapsed=33:00:38 avg_step=1.34s recent_step=1.14s eta=04:09:18 +2026-08-26 12:56:21 | INFO | train rank=0 | step=88830/100000 progress=88.8% loss=0.069427 lr=3.372525e-07 elapsed=33:00:51 avg_step=1.34s recent_step=1.38s eta=04:09:05 +2026-08-26 12:56:33 | INFO | train rank=0 | step=88840/100000 progress=88.8% loss=0.037476 lr=3.366558e-07 elapsed=33:01:03 avg_step=1.34s recent_step=1.15s eta=04:08:51 +2026-08-26 12:56:45 | INFO | train rank=0 | step=88850/100000 progress=88.8% loss=0.035400 lr=3.360595e-07 elapsed=33:01:15 avg_step=1.34s recent_step=1.23s eta=04:08:37 +2026-08-26 12:57:01 | INFO | train rank=0 | step=88860/100000 progress=88.9% loss=0.064331 lr=3.354638e-07 elapsed=33:01:30 avg_step=1.34s recent_step=1.53s eta=04:08:24 +2026-08-26 12:57:16 | INFO | train rank=0 | step=88870/100000 progress=88.9% loss=0.037621 lr=3.348687e-07 elapsed=33:01:45 avg_step=1.34s recent_step=1.50s eta=04:08:11 +2026-08-26 12:57:27 | INFO | train rank=0 | step=88880/100000 progress=88.9% loss=0.038414 lr=3.342740e-07 elapsed=33:01:57 avg_step=1.34s recent_step=1.18s eta=04:07:58 +2026-08-26 12:57:39 | INFO | train rank=0 | step=88890/100000 progress=88.9% loss=0.048630 lr=3.336798e-07 elapsed=33:02:09 avg_step=1.34s recent_step=1.20s eta=04:07:44 +2026-08-26 12:57:53 | INFO | train rank=0 | step=88900/100000 progress=88.9% loss=0.043877 lr=3.330862e-07 elapsed=33:02:23 avg_step=1.34s recent_step=1.36s eta=04:07:31 +2026-08-26 12:58:09 | INFO | train rank=0 | step=88910/100000 progress=88.9% loss=0.053101 lr=3.324930e-07 elapsed=33:02:39 avg_step=1.34s recent_step=1.65s eta=04:07:18 +2026-08-26 12:58:21 | INFO | train rank=0 | step=88920/100000 progress=88.9% loss=0.050858 lr=3.319004e-07 elapsed=33:02:51 avg_step=1.34s recent_step=1.17s eta=04:07:04 +2026-08-26 12:58:39 | INFO | train rank=0 | step=88930/100000 progress=88.9% loss=0.039871 lr=3.313083e-07 elapsed=33:03:09 avg_step=1.34s recent_step=1.76s eta=04:06:51 +2026-08-26 12:58:50 | INFO | train rank=0 | step=88940/100000 progress=88.9% loss=0.046570 lr=3.307166e-07 elapsed=33:03:20 avg_step=1.34s recent_step=1.18s eta=04:06:38 +2026-08-26 12:59:02 | INFO | train rank=0 | step=88950/100000 progress=88.9% loss=0.037231 lr=3.301255e-07 elapsed=33:03:32 avg_step=1.34s recent_step=1.16s eta=04:06:24 +2026-08-26 12:59:14 | INFO | train rank=0 | step=88960/100000 progress=89.0% loss=0.027828 lr=3.295349e-07 elapsed=33:03:44 avg_step=1.34s recent_step=1.21s eta=04:06:11 +2026-08-26 12:59:27 | INFO | train rank=0 | step=88970/100000 progress=89.0% loss=0.061890 lr=3.289449e-07 elapsed=33:03:57 avg_step=1.34s recent_step=1.32s eta=04:05:57 +2026-08-26 12:59:42 | INFO | train rank=0 | step=88980/100000 progress=89.0% loss=0.051224 lr=3.283553e-07 elapsed=33:04:12 avg_step=1.34s recent_step=1.46s eta=04:05:44 +2026-08-26 12:59:56 | INFO | train rank=0 | step=88990/100000 progress=89.0% loss=0.035759 lr=3.277662e-07 elapsed=33:04:26 avg_step=1.34s recent_step=1.39s eta=04:05:31 +2026-08-26 13:00:11 | INFO | train rank=0 | step=89000/100000 progress=89.0% loss=0.028038 lr=3.271777e-07 elapsed=33:04:41 avg_step=1.34s recent_step=1.50s eta=04:05:17 +2026-08-26 13:00:23 | INFO | train rank=0 | step=89010/100000 progress=89.0% loss=0.032028 lr=3.265896e-07 elapsed=33:04:53 avg_step=1.34s recent_step=1.22s eta=04:05:04 +2026-08-26 13:00:36 | INFO | train rank=0 | step=89020/100000 progress=89.0% loss=0.037018 lr=3.260021e-07 elapsed=33:05:06 avg_step=1.34s recent_step=1.32s eta=04:04:50 +2026-08-26 13:00:49 | INFO | train rank=0 | step=89030/100000 progress=89.0% loss=0.039642 lr=3.254151e-07 elapsed=33:05:18 avg_step=1.34s recent_step=1.22s eta=04:04:37 +2026-08-26 13:01:01 | INFO | train rank=0 | step=89040/100000 progress=89.0% loss=0.043312 lr=3.248286e-07 elapsed=33:05:30 avg_step=1.34s recent_step=1.20s eta=04:04:23 +2026-08-26 13:01:15 | INFO | train rank=0 | step=89050/100000 progress=89.0% loss=0.050720 lr=3.242426e-07 elapsed=33:05:45 avg_step=1.34s recent_step=1.48s eta=04:04:10 +2026-08-26 13:01:30 | INFO | train rank=0 | step=89060/100000 progress=89.1% loss=0.046860 lr=3.236571e-07 elapsed=33:06:00 avg_step=1.34s recent_step=1.47s eta=04:03:57 +2026-08-26 13:01:42 | INFO | train rank=0 | step=89070/100000 progress=89.1% loss=0.062744 lr=3.230721e-07 elapsed=33:06:12 avg_step=1.34s recent_step=1.19s eta=04:03:43 +2026-08-26 13:01:54 | INFO | train rank=0 | step=89080/100000 progress=89.1% loss=0.020706 lr=3.224877e-07 elapsed=33:06:24 avg_step=1.34s recent_step=1.23s eta=04:03:30 +2026-08-26 13:02:07 | INFO | train rank=0 | step=89090/100000 progress=89.1% loss=0.034332 lr=3.219037e-07 elapsed=33:06:36 avg_step=1.34s recent_step=1.23s eta=04:03:16 +2026-08-26 13:02:18 | INFO | train rank=0 | step=89100/100000 progress=89.1% loss=0.047821 lr=3.213203e-07 elapsed=33:06:48 avg_step=1.34s recent_step=1.19s eta=04:03:03 +2026-08-26 13:02:36 | INFO | train rank=0 | step=89110/100000 progress=89.1% loss=0.044846 lr=3.207374e-07 elapsed=33:07:06 avg_step=1.34s recent_step=1.75s eta=04:02:50 +2026-08-26 13:02:48 | INFO | train rank=0 | step=89120/100000 progress=89.1% loss=0.050110 lr=3.201550e-07 elapsed=33:07:18 avg_step=1.34s recent_step=1.20s eta=04:02:36 +2026-08-26 13:03:00 | INFO | train rank=0 | step=89130/100000 progress=89.1% loss=0.102020 lr=3.195731e-07 elapsed=33:07:30 avg_step=1.34s recent_step=1.19s eta=04:02:23 +2026-08-26 13:03:12 | INFO | train rank=0 | step=89140/100000 progress=89.1% loss=0.040909 lr=3.189917e-07 elapsed=33:07:42 avg_step=1.34s recent_step=1.20s eta=04:02:09 +2026-08-26 13:03:24 | INFO | train rank=0 | step=89150/100000 progress=89.1% loss=0.031250 lr=3.184108e-07 elapsed=33:07:54 avg_step=1.34s recent_step=1.25s eta=04:01:56 +2026-08-26 13:03:37 | INFO | train rank=0 | step=89160/100000 progress=89.2% loss=0.040886 lr=3.178304e-07 elapsed=33:08:07 avg_step=1.34s recent_step=1.30s eta=04:01:42 +2026-08-26 13:03:51 | INFO | train rank=0 | step=89170/100000 progress=89.2% loss=0.037735 lr=3.172506e-07 elapsed=33:08:21 avg_step=1.34s recent_step=1.35s eta=04:01:29 +2026-08-26 13:04:04 | INFO | train rank=0 | step=89180/100000 progress=89.2% loss=0.045609 lr=3.166712e-07 elapsed=33:08:34 avg_step=1.34s recent_step=1.28s eta=04:01:16 +2026-08-26 13:04:20 | INFO | train rank=0 | step=89190/100000 progress=89.2% loss=0.043655 lr=3.160924e-07 elapsed=33:08:50 avg_step=1.34s recent_step=1.61s eta=04:01:03 +2026-08-26 13:04:32 | INFO | train rank=0 | step=89200/100000 progress=89.2% loss=0.048660 lr=3.155141e-07 elapsed=33:09:02 avg_step=1.34s recent_step=1.27s eta=04:00:49 +2026-08-26 13:04:48 | INFO | train rank=0 | step=89210/100000 progress=89.2% loss=0.055183 lr=3.149363e-07 elapsed=33:09:18 avg_step=1.34s recent_step=1.55s eta=04:00:36 +2026-08-26 13:05:00 | INFO | train rank=0 | step=89220/100000 progress=89.2% loss=0.053711 lr=3.143590e-07 elapsed=33:09:30 avg_step=1.34s recent_step=1.18s eta=04:00:22 +2026-08-26 13:05:14 | INFO | train rank=0 | step=89230/100000 progress=89.2% loss=0.046951 lr=3.137822e-07 elapsed=33:09:44 avg_step=1.34s recent_step=1.40s eta=04:00:09 +2026-08-26 13:05:28 | INFO | train rank=0 | step=89240/100000 progress=89.2% loss=0.026237 lr=3.132059e-07 elapsed=33:09:58 avg_step=1.34s recent_step=1.43s eta=03:59:56 +2026-08-26 13:05:40 | INFO | train rank=0 | step=89250/100000 progress=89.2% loss=0.063053 lr=3.126302e-07 elapsed=33:10:10 avg_step=1.34s recent_step=1.18s eta=03:59:42 +2026-08-26 13:05:55 | INFO | train rank=0 | step=89260/100000 progress=89.3% loss=0.039268 lr=3.120549e-07 elapsed=33:10:25 avg_step=1.34s recent_step=1.50s eta=03:59:29 +2026-08-26 13:06:10 | INFO | train rank=0 | step=89270/100000 progress=89.3% loss=0.042450 lr=3.114802e-07 elapsed=33:10:40 avg_step=1.34s recent_step=1.57s eta=03:59:16 +2026-08-26 13:06:24 | INFO | train rank=0 | step=89280/100000 progress=89.3% loss=0.038727 lr=3.109060e-07 elapsed=33:10:54 avg_step=1.34s recent_step=1.37s eta=03:59:03 +2026-08-26 13:06:37 | INFO | train rank=0 | step=89290/100000 progress=89.3% loss=0.061981 lr=3.103323e-07 elapsed=33:11:07 avg_step=1.34s recent_step=1.25s eta=03:58:49 +2026-08-26 13:06:48 | INFO | train rank=0 | step=89300/100000 progress=89.3% loss=0.040268 lr=3.097591e-07 elapsed=33:11:18 avg_step=1.34s recent_step=1.18s eta=03:58:36 +2026-08-26 13:07:00 | INFO | train rank=0 | step=89310/100000 progress=89.3% loss=0.076691 lr=3.091864e-07 elapsed=33:11:30 avg_step=1.34s recent_step=1.17s eta=03:58:22 +2026-08-26 13:07:12 | INFO | train rank=0 | step=89320/100000 progress=89.3% loss=0.026741 lr=3.086142e-07 elapsed=33:11:42 avg_step=1.34s recent_step=1.17s eta=03:58:08 +2026-08-26 13:07:25 | INFO | train rank=0 | step=89330/100000 progress=89.3% loss=0.046265 lr=3.080426e-07 elapsed=33:11:54 avg_step=1.34s recent_step=1.28s eta=03:57:55 +2026-08-26 13:07:39 | INFO | train rank=0 | step=89340/100000 progress=89.3% loss=0.035339 lr=3.074715e-07 elapsed=33:12:09 avg_step=1.34s recent_step=1.47s eta=03:57:42 +2026-08-26 13:07:55 | INFO | train rank=0 | step=89350/100000 progress=89.3% loss=0.027298 lr=3.069008e-07 elapsed=33:12:25 avg_step=1.34s recent_step=1.57s eta=03:57:29 +2026-08-26 13:08:07 | INFO | train rank=0 | step=89360/100000 progress=89.4% loss=0.032166 lr=3.063307e-07 elapsed=33:12:37 avg_step=1.34s recent_step=1.19s eta=03:57:15 +2026-08-26 13:08:20 | INFO | train rank=0 | step=89370/100000 progress=89.4% loss=0.049133 lr=3.057611e-07 elapsed=33:12:50 avg_step=1.34s recent_step=1.30s eta=03:57:02 +2026-08-26 13:08:34 | INFO | train rank=0 | step=89380/100000 progress=89.4% loss=0.039314 lr=3.051920e-07 elapsed=33:13:04 avg_step=1.34s recent_step=1.46s eta=03:56:48 +2026-08-26 13:08:50 | INFO | train rank=0 | step=89390/100000 progress=89.4% loss=0.041351 lr=3.046235e-07 elapsed=33:13:20 avg_step=1.34s recent_step=1.54s eta=03:56:35 +2026-08-26 13:09:05 | INFO | train rank=0 | step=89400/100000 progress=89.4% loss=0.033737 lr=3.040554e-07 elapsed=33:13:35 avg_step=1.34s recent_step=1.54s eta=03:56:22 +2026-08-26 13:09:17 | INFO | train rank=0 | step=89410/100000 progress=89.4% loss=0.027596 lr=3.034878e-07 elapsed=33:13:47 avg_step=1.34s recent_step=1.21s eta=03:56:09 +2026-08-26 13:09:29 | INFO | train rank=0 | step=89420/100000 progress=89.4% loss=0.043762 lr=3.029208e-07 elapsed=33:13:59 avg_step=1.34s recent_step=1.16s eta=03:55:55 +2026-08-26 13:09:43 | INFO | train rank=0 | step=89430/100000 progress=89.4% loss=0.054962 lr=3.023543e-07 elapsed=33:14:13 avg_step=1.34s recent_step=1.40s eta=03:55:42 +2026-08-26 13:09:56 | INFO | train rank=0 | step=89440/100000 progress=89.4% loss=0.045563 lr=3.017883e-07 elapsed=33:14:26 avg_step=1.34s recent_step=1.34s eta=03:55:28 +2026-08-26 13:10:09 | INFO | train rank=0 | step=89450/100000 progress=89.5% loss=0.062485 lr=3.012228e-07 elapsed=33:14:39 avg_step=1.34s recent_step=1.26s eta=03:55:15 +2026-08-26 13:10:21 | INFO | train rank=0 | step=89460/100000 progress=89.5% loss=0.071777 lr=3.006578e-07 elapsed=33:14:51 avg_step=1.34s recent_step=1.20s eta=03:55:01 +2026-08-26 13:10:35 | INFO | train rank=0 | step=89470/100000 progress=89.5% loss=0.043510 lr=3.000934e-07 elapsed=33:15:04 avg_step=1.34s recent_step=1.36s eta=03:54:48 +2026-08-26 13:10:47 | INFO | train rank=0 | step=89480/100000 progress=89.5% loss=0.042786 lr=2.995294e-07 elapsed=33:15:17 avg_step=1.34s recent_step=1.23s eta=03:54:34 +2026-08-26 13:10:59 | INFO | train rank=0 | step=89490/100000 progress=89.5% loss=0.026840 lr=2.989660e-07 elapsed=33:15:29 avg_step=1.34s recent_step=1.26s eta=03:54:21 +2026-08-26 13:11:18 | INFO | train rank=0 | step=89500/100000 progress=89.5% loss=0.048523 lr=2.984030e-07 elapsed=33:15:48 avg_step=1.34s recent_step=1.87s eta=03:54:08 +2026-08-26 13:11:30 | INFO | train rank=0 | step=89510/100000 progress=89.5% loss=0.039230 lr=2.978406e-07 elapsed=33:16:00 avg_step=1.34s recent_step=1.19s eta=03:53:55 +2026-08-26 13:11:45 | INFO | train rank=0 | step=89520/100000 progress=89.5% loss=0.039734 lr=2.972787e-07 elapsed=33:16:15 avg_step=1.34s recent_step=1.51s eta=03:53:41 +2026-08-26 13:11:58 | INFO | train rank=0 | step=89530/100000 progress=89.5% loss=0.056335 lr=2.967174e-07 elapsed=33:16:27 avg_step=1.34s recent_step=1.24s eta=03:53:28 +2026-08-26 13:12:12 | INFO | train rank=0 | step=89540/100000 progress=89.5% loss=0.043167 lr=2.961565e-07 elapsed=33:16:41 avg_step=1.34s recent_step=1.41s eta=03:53:15 +2026-08-26 13:12:27 | INFO | train rank=0 | step=89550/100000 progress=89.5% loss=0.046448 lr=2.955962e-07 elapsed=33:16:57 avg_step=1.34s recent_step=1.54s eta=03:53:02 +2026-08-26 13:12:41 | INFO | train rank=0 | step=89560/100000 progress=89.6% loss=0.046860 lr=2.950363e-07 elapsed=33:17:11 avg_step=1.34s recent_step=1.42s eta=03:52:48 +2026-08-26 13:12:55 | INFO | train rank=0 | step=89570/100000 progress=89.6% loss=0.044067 lr=2.944770e-07 elapsed=33:17:24 avg_step=1.34s recent_step=1.34s eta=03:52:35 +2026-08-26 13:13:07 | INFO | train rank=0 | step=89580/100000 progress=89.6% loss=0.038010 lr=2.939182e-07 elapsed=33:17:36 avg_step=1.34s recent_step=1.20s eta=03:52:21 +2026-08-26 13:13:22 | INFO | train rank=0 | step=89590/100000 progress=89.6% loss=0.046875 lr=2.933599e-07 elapsed=33:17:52 avg_step=1.34s recent_step=1.57s eta=03:52:08 +2026-08-26 13:13:34 | INFO | train rank=0 | step=89600/100000 progress=89.6% loss=0.061676 lr=2.928021e-07 elapsed=33:18:04 avg_step=1.34s recent_step=1.19s eta=03:51:55 +2026-08-26 13:13:47 | INFO | train rank=0 | step=89610/100000 progress=89.6% loss=0.068115 lr=2.922449e-07 elapsed=33:18:17 avg_step=1.34s recent_step=1.27s eta=03:51:41 +2026-08-26 13:14:00 | INFO | train rank=0 | step=89620/100000 progress=89.6% loss=0.071075 lr=2.916881e-07 elapsed=33:18:30 avg_step=1.34s recent_step=1.33s eta=03:51:28 +2026-08-26 13:14:12 | INFO | train rank=0 | step=89630/100000 progress=89.6% loss=0.033218 lr=2.911319e-07 elapsed=33:18:41 avg_step=1.34s recent_step=1.15s eta=03:51:14 +2026-08-26 13:14:24 | INFO | train rank=0 | step=89640/100000 progress=89.6% loss=0.035126 lr=2.905762e-07 elapsed=33:18:54 avg_step=1.34s recent_step=1.24s eta=03:51:01 +2026-08-26 13:14:40 | INFO | train rank=0 | step=89650/100000 progress=89.6% loss=0.024719 lr=2.900210e-07 elapsed=33:19:10 avg_step=1.34s recent_step=1.61s eta=03:50:48 +2026-08-26 13:14:53 | INFO | train rank=0 | step=89660/100000 progress=89.7% loss=0.036835 lr=2.894663e-07 elapsed=33:19:23 avg_step=1.34s recent_step=1.35s eta=03:50:34 +2026-08-26 13:15:08 | INFO | train rank=0 | step=89670/100000 progress=89.7% loss=0.049271 lr=2.889121e-07 elapsed=33:19:38 avg_step=1.34s recent_step=1.47s eta=03:50:21 +2026-08-26 13:15:21 | INFO | train rank=0 | step=89680/100000 progress=89.7% loss=0.067627 lr=2.883584e-07 elapsed=33:19:51 avg_step=1.34s recent_step=1.26s eta=03:50:08 +2026-08-26 13:15:35 | INFO | train rank=0 | step=89690/100000 progress=89.7% loss=0.054329 lr=2.878053e-07 elapsed=33:20:05 avg_step=1.34s recent_step=1.40s eta=03:49:54 +2026-08-26 13:15:48 | INFO | train rank=0 | step=89700/100000 progress=89.7% loss=0.108887 lr=2.872527e-07 elapsed=33:20:18 avg_step=1.34s recent_step=1.32s eta=03:49:41 +2026-08-26 13:16:02 | INFO | train rank=0 | step=89710/100000 progress=89.7% loss=0.071564 lr=2.867006e-07 elapsed=33:20:32 avg_step=1.34s recent_step=1.38s eta=03:49:28 +2026-08-26 13:16:15 | INFO | train rank=0 | step=89720/100000 progress=89.7% loss=0.054367 lr=2.861490e-07 elapsed=33:20:45 avg_step=1.34s recent_step=1.30s eta=03:49:14 +2026-08-26 13:16:30 | INFO | train rank=0 | step=89730/100000 progress=89.7% loss=0.039703 lr=2.855979e-07 elapsed=33:21:00 avg_step=1.34s recent_step=1.51s eta=03:49:01 +2026-08-26 13:16:46 | INFO | train rank=0 | step=89740/100000 progress=89.7% loss=0.039043 lr=2.850473e-07 elapsed=33:21:16 avg_step=1.34s recent_step=1.58s eta=03:48:48 +2026-08-26 13:16:58 | INFO | train rank=0 | step=89750/100000 progress=89.8% loss=0.052673 lr=2.844973e-07 elapsed=33:21:28 avg_step=1.34s recent_step=1.21s eta=03:48:34 +2026-08-26 13:17:09 | INFO | train rank=0 | step=89760/100000 progress=89.8% loss=0.039146 lr=2.839477e-07 elapsed=33:21:39 avg_step=1.34s recent_step=1.16s eta=03:48:21 +2026-08-26 13:17:21 | INFO | train rank=0 | step=89770/100000 progress=89.8% loss=0.051659 lr=2.833987e-07 elapsed=33:21:51 avg_step=1.34s recent_step=1.19s eta=03:48:07 +2026-08-26 13:17:35 | INFO | train rank=0 | step=89780/100000 progress=89.8% loss=0.083115 lr=2.828502e-07 elapsed=33:22:05 avg_step=1.34s recent_step=1.42s eta=03:47:54 +2026-08-26 13:17:50 | INFO | train rank=0 | step=89790/100000 progress=89.8% loss=0.045120 lr=2.823022e-07 elapsed=33:22:20 avg_step=1.34s recent_step=1.48s eta=03:47:41 +2026-08-26 13:18:02 | INFO | train rank=0 | step=89800/100000 progress=89.8% loss=0.045082 lr=2.817548e-07 elapsed=33:22:32 avg_step=1.34s recent_step=1.17s eta=03:47:27 +2026-08-26 13:18:14 | INFO | train rank=0 | step=89810/100000 progress=89.8% loss=0.048325 lr=2.812078e-07 elapsed=33:22:43 avg_step=1.34s recent_step=1.18s eta=03:47:14 +2026-08-26 13:18:30 | INFO | train rank=0 | step=89820/100000 progress=89.8% loss=0.042542 lr=2.806614e-07 elapsed=33:23:00 avg_step=1.34s recent_step=1.64s eta=03:47:00 +2026-08-26 13:18:42 | INFO | train rank=0 | step=89830/100000 progress=89.8% loss=0.037659 lr=2.801154e-07 elapsed=33:23:12 avg_step=1.34s recent_step=1.19s eta=03:46:47 +2026-08-26 13:18:53 | INFO | train rank=0 | step=89840/100000 progress=89.8% loss=0.068481 lr=2.795700e-07 elapsed=33:23:23 avg_step=1.34s recent_step=1.14s eta=03:46:33 +2026-08-26 13:19:06 | INFO | train rank=0 | step=89850/100000 progress=89.8% loss=0.026581 lr=2.790252e-07 elapsed=33:23:36 avg_step=1.34s recent_step=1.23s eta=03:46:20 +2026-08-26 13:19:21 | INFO | train rank=0 | step=89860/100000 progress=89.9% loss=0.054947 lr=2.784808e-07 elapsed=33:23:51 avg_step=1.34s recent_step=1.59s eta=03:46:07 +2026-08-26 13:19:34 | INFO | train rank=0 | step=89870/100000 progress=89.9% loss=0.025665 lr=2.779369e-07 elapsed=33:24:04 avg_step=1.34s recent_step=1.23s eta=03:45:53 +2026-08-26 13:19:48 | INFO | train rank=0 | step=89880/100000 progress=89.9% loss=0.071411 lr=2.773936e-07 elapsed=33:24:18 avg_step=1.34s recent_step=1.46s eta=03:45:40 +2026-08-26 13:20:00 | INFO | train rank=0 | step=89890/100000 progress=89.9% loss=0.042992 lr=2.768508e-07 elapsed=33:24:30 avg_step=1.34s recent_step=1.15s eta=03:45:26 +2026-08-26 13:20:17 | INFO | train rank=0 | step=89900/100000 progress=89.9% loss=0.061295 lr=2.763084e-07 elapsed=33:24:47 avg_step=1.34s recent_step=1.72s eta=03:45:13 +2026-08-26 13:20:29 | INFO | train rank=0 | step=89910/100000 progress=89.9% loss=0.040939 lr=2.757667e-07 elapsed=33:24:59 avg_step=1.34s recent_step=1.17s eta=03:45:00 +2026-08-26 13:20:40 | INFO | train rank=0 | step=89920/100000 progress=89.9% loss=0.067322 lr=2.752254e-07 elapsed=33:25:10 avg_step=1.34s recent_step=1.16s eta=03:44:46 +2026-08-26 13:20:52 | INFO | train rank=0 | step=89930/100000 progress=89.9% loss=0.043701 lr=2.746846e-07 elapsed=33:25:22 avg_step=1.34s recent_step=1.19s eta=03:44:33 +2026-08-26 13:21:06 | INFO | train rank=0 | step=89940/100000 progress=89.9% loss=0.039780 lr=2.741444e-07 elapsed=33:25:36 avg_step=1.34s recent_step=1.38s eta=03:44:19 +2026-08-26 13:21:22 | INFO | train rank=0 | step=89950/100000 progress=90.0% loss=0.033005 lr=2.736047e-07 elapsed=33:25:52 avg_step=1.34s recent_step=1.63s eta=03:44:06 +2026-08-26 13:21:34 | INFO | train rank=0 | step=89960/100000 progress=90.0% loss=0.055328 lr=2.730655e-07 elapsed=33:26:04 avg_step=1.34s recent_step=1.20s eta=03:43:53 +2026-08-26 13:21:46 | INFO | train rank=0 | step=89970/100000 progress=90.0% loss=0.049316 lr=2.725268e-07 elapsed=33:26:16 avg_step=1.34s recent_step=1.16s eta=03:43:39 +2026-08-26 13:22:03 | INFO | train rank=0 | step=89980/100000 progress=90.0% loss=0.049690 lr=2.719886e-07 elapsed=33:26:33 avg_step=1.34s recent_step=1.70s eta=03:43:26 +2026-08-26 13:22:15 | INFO | train rank=0 | step=89990/100000 progress=90.0% loss=0.055405 lr=2.714509e-07 elapsed=33:26:45 avg_step=1.34s recent_step=1.18s eta=03:43:13 +2026-08-26 13:22:31 | INFO | train rank=0 | step=90000/100000 progress=90.0% loss=0.039810 lr=2.709138e-07 elapsed=33:27:01 avg_step=1.34s recent_step=1.62s eta=03:43:00 +2026-08-26 13:22:42 | INFO | train rank=0 | step=90010/100000 progress=90.0% loss=0.056366 lr=2.703772e-07 elapsed=33:27:12 avg_step=1.34s recent_step=1.14s eta=03:42:46 +2026-08-26 13:22:54 | INFO | train rank=0 | step=90020/100000 progress=90.0% loss=0.046371 lr=2.698411e-07 elapsed=33:27:23 avg_step=1.34s recent_step=1.12s eta=03:42:32 +2026-08-26 13:23:07 | INFO | train rank=0 | step=90030/100000 progress=90.0% loss=0.062119 lr=2.693055e-07 elapsed=33:27:37 avg_step=1.34s recent_step=1.34s eta=03:42:19 +2026-08-26 13:23:19 | INFO | train rank=0 | step=90040/100000 progress=90.0% loss=0.039642 lr=2.687704e-07 elapsed=33:27:49 avg_step=1.34s recent_step=1.17s eta=03:42:05 +2026-08-26 13:23:34 | INFO | train rank=0 | step=90050/100000 progress=90.0% loss=0.053818 lr=2.682359e-07 elapsed=33:28:04 avg_step=1.34s recent_step=1.58s eta=03:41:52 +2026-08-26 13:23:48 | INFO | train rank=0 | step=90060/100000 progress=90.1% loss=0.032181 lr=2.677018e-07 elapsed=33:28:18 avg_step=1.34s recent_step=1.35s eta=03:41:39 +2026-08-26 13:24:03 | INFO | train rank=0 | step=90070/100000 progress=90.1% loss=0.025398 lr=2.671683e-07 elapsed=33:28:33 avg_step=1.34s recent_step=1.49s eta=03:41:26 +2026-08-26 13:24:18 | INFO | train rank=0 | step=90080/100000 progress=90.1% loss=0.104309 lr=2.666353e-07 elapsed=33:28:48 avg_step=1.34s recent_step=1.51s eta=03:41:13 +2026-08-26 13:24:30 | INFO | train rank=0 | step=90090/100000 progress=90.1% loss=0.042450 lr=2.661028e-07 elapsed=33:29:00 avg_step=1.34s recent_step=1.20s eta=03:40:59 +2026-08-26 13:24:48 | INFO | train rank=0 | step=90100/100000 progress=90.1% loss=0.035118 lr=2.655709e-07 elapsed=33:29:18 avg_step=1.34s recent_step=1.80s eta=03:40:46 +2026-08-26 13:25:00 | INFO | train rank=0 | step=90110/100000 progress=90.1% loss=0.042625 lr=2.650394e-07 elapsed=33:29:30 avg_step=1.34s recent_step=1.21s eta=03:40:33 +2026-08-26 13:25:13 | INFO | train rank=0 | step=90120/100000 progress=90.1% loss=0.089813 lr=2.645085e-07 elapsed=33:29:43 avg_step=1.34s recent_step=1.28s eta=03:40:19 +2026-08-26 13:25:25 | INFO | train rank=0 | step=90130/100000 progress=90.1% loss=0.044785 lr=2.639781e-07 elapsed=33:29:55 avg_step=1.34s recent_step=1.22s eta=03:40:06 +2026-08-26 13:25:37 | INFO | train rank=0 | step=90140/100000 progress=90.1% loss=0.046173 lr=2.634482e-07 elapsed=33:30:07 avg_step=1.34s recent_step=1.15s eta=03:39:52 +2026-08-26 13:25:52 | INFO | train rank=0 | step=90150/100000 progress=90.1% loss=0.036240 lr=2.629188e-07 elapsed=33:30:22 avg_step=1.34s recent_step=1.52s eta=03:39:39 +2026-08-26 13:26:04 | INFO | train rank=0 | step=90160/100000 progress=90.2% loss=0.034630 lr=2.623899e-07 elapsed=33:30:34 avg_step=1.34s recent_step=1.18s eta=03:39:25 +2026-08-26 13:26:16 | INFO | train rank=0 | step=90170/100000 progress=90.2% loss=0.051086 lr=2.618616e-07 elapsed=33:30:46 avg_step=1.34s recent_step=1.23s eta=03:39:12 +2026-08-26 13:26:34 | INFO | train rank=0 | step=90180/100000 progress=90.2% loss=0.059738 lr=2.613338e-07 elapsed=33:31:04 avg_step=1.34s recent_step=1.77s eta=03:38:59 +2026-08-26 13:26:45 | INFO | train rank=0 | step=90190/100000 progress=90.2% loss=0.062531 lr=2.608065e-07 elapsed=33:31:15 avg_step=1.34s recent_step=1.16s eta=03:38:45 +2026-08-26 13:26:57 | INFO | train rank=0 | step=90200/100000 progress=90.2% loss=0.038544 lr=2.602797e-07 elapsed=33:31:27 avg_step=1.34s recent_step=1.15s eta=03:38:32 +2026-08-26 13:27:08 | INFO | train rank=0 | step=90210/100000 progress=90.2% loss=0.080872 lr=2.597534e-07 elapsed=33:31:38 avg_step=1.34s recent_step=1.15s eta=03:38:18 +2026-08-26 13:27:23 | INFO | train rank=0 | step=90220/100000 progress=90.2% loss=0.066071 lr=2.592277e-07 elapsed=33:31:53 avg_step=1.34s recent_step=1.49s eta=03:38:05 +2026-08-26 13:27:35 | INFO | train rank=0 | step=90230/100000 progress=90.2% loss=0.044189 lr=2.587024e-07 elapsed=33:32:05 avg_step=1.34s recent_step=1.20s eta=03:37:52 +2026-08-26 13:27:50 | INFO | train rank=0 | step=90240/100000 progress=90.2% loss=0.069214 lr=2.581777e-07 elapsed=33:32:20 avg_step=1.34s recent_step=1.44s eta=03:37:38 +2026-08-26 13:28:05 | INFO | train rank=0 | step=90250/100000 progress=90.2% loss=0.031624 lr=2.576535e-07 elapsed=33:32:35 avg_step=1.34s recent_step=1.51s eta=03:37:25 +2026-08-26 13:28:18 | INFO | train rank=0 | step=90260/100000 progress=90.3% loss=0.064941 lr=2.571299e-07 elapsed=33:32:48 avg_step=1.34s recent_step=1.38s eta=03:37:12 +2026-08-26 13:28:30 | INFO | train rank=0 | step=90270/100000 progress=90.3% loss=0.051422 lr=2.566067e-07 elapsed=33:33:00 avg_step=1.34s recent_step=1.13s eta=03:36:58 +2026-08-26 13:28:41 | INFO | train rank=0 | step=90280/100000 progress=90.3% loss=0.033920 lr=2.560841e-07 elapsed=33:33:11 avg_step=1.34s recent_step=1.13s eta=03:36:45 +2026-08-26 13:28:58 | INFO | train rank=0 | step=90290/100000 progress=90.3% loss=0.072754 lr=2.555620e-07 elapsed=33:33:28 avg_step=1.34s recent_step=1.65s eta=03:36:31 +2026-08-26 13:29:10 | INFO | train rank=0 | step=90300/100000 progress=90.3% loss=0.054474 lr=2.550404e-07 elapsed=33:33:39 avg_step=1.34s recent_step=1.19s eta=03:36:18 +2026-08-26 13:29:21 | INFO | train rank=0 | step=90310/100000 progress=90.3% loss=0.048630 lr=2.545193e-07 elapsed=33:33:51 avg_step=1.34s recent_step=1.12s eta=03:36:04 +2026-08-26 13:29:35 | INFO | train rank=0 | step=90320/100000 progress=90.3% loss=0.065796 lr=2.539987e-07 elapsed=33:34:05 avg_step=1.34s recent_step=1.40s eta=03:35:51 +2026-08-26 13:29:49 | INFO | train rank=0 | step=90330/100000 progress=90.3% loss=0.045212 lr=2.534787e-07 elapsed=33:34:19 avg_step=1.34s recent_step=1.47s eta=03:35:38 +2026-08-26 13:30:06 | INFO | train rank=0 | step=90340/100000 progress=90.3% loss=0.044525 lr=2.529591e-07 elapsed=33:34:36 avg_step=1.34s recent_step=1.68s eta=03:35:25 +2026-08-26 13:30:18 | INFO | train rank=0 | step=90350/100000 progress=90.3% loss=0.050674 lr=2.524401e-07 elapsed=33:34:48 avg_step=1.34s recent_step=1.19s eta=03:35:11 +2026-08-26 13:30:35 | INFO | train rank=0 | step=90360/100000 progress=90.4% loss=0.048981 lr=2.519217e-07 elapsed=33:35:05 avg_step=1.34s recent_step=1.72s eta=03:34:58 +2026-08-26 13:30:47 | INFO | train rank=0 | step=90370/100000 progress=90.4% loss=0.042557 lr=2.514037e-07 elapsed=33:35:17 avg_step=1.34s recent_step=1.19s eta=03:34:45 +2026-08-26 13:31:02 | INFO | train rank=0 | step=90380/100000 progress=90.4% loss=0.055878 lr=2.508862e-07 elapsed=33:35:32 avg_step=1.34s recent_step=1.50s eta=03:34:31 +2026-08-26 13:31:18 | INFO | train rank=0 | step=90390/100000 progress=90.4% loss=0.038101 lr=2.503693e-07 elapsed=33:35:48 avg_step=1.34s recent_step=1.56s eta=03:34:18 +2026-08-26 13:31:29 | INFO | train rank=0 | step=90400/100000 progress=90.4% loss=0.028267 lr=2.498529e-07 elapsed=33:35:59 avg_step=1.34s recent_step=1.16s eta=03:34:05 +2026-08-26 13:31:43 | INFO | train rank=0 | step=90410/100000 progress=90.4% loss=0.060760 lr=2.493370e-07 elapsed=33:36:13 avg_step=1.34s recent_step=1.36s eta=03:33:51 +2026-08-26 13:31:57 | INFO | train rank=0 | step=90420/100000 progress=90.4% loss=0.044586 lr=2.488217e-07 elapsed=33:36:27 avg_step=1.34s recent_step=1.42s eta=03:33:38 +2026-08-26 13:32:08 | INFO | train rank=0 | step=90430/100000 progress=90.4% loss=0.047684 lr=2.483068e-07 elapsed=33:36:38 avg_step=1.34s recent_step=1.14s eta=03:33:25 +2026-08-26 13:32:23 | INFO | train rank=0 | step=90440/100000 progress=90.4% loss=0.048416 lr=2.477925e-07 elapsed=33:36:53 avg_step=1.34s recent_step=1.43s eta=03:33:11 +2026-08-26 13:32:36 | INFO | train rank=0 | step=90450/100000 progress=90.5% loss=0.054962 lr=2.472787e-07 elapsed=33:37:06 avg_step=1.34s recent_step=1.33s eta=03:32:58 +2026-08-26 13:32:53 | INFO | train rank=0 | step=90460/100000 progress=90.5% loss=0.032379 lr=2.467654e-07 elapsed=33:37:23 avg_step=1.34s recent_step=1.69s eta=03:32:45 +2026-08-26 13:33:06 | INFO | train rank=0 | step=90470/100000 progress=90.5% loss=0.107910 lr=2.462526e-07 elapsed=33:37:36 avg_step=1.34s recent_step=1.34s eta=03:32:31 +2026-08-26 13:33:18 | INFO | train rank=0 | step=90480/100000 progress=90.5% loss=0.035637 lr=2.457404e-07 elapsed=33:37:48 avg_step=1.34s recent_step=1.16s eta=03:32:18 +2026-08-26 13:33:33 | INFO | train rank=0 | step=90490/100000 progress=90.5% loss=0.068481 lr=2.452286e-07 elapsed=33:38:02 avg_step=1.34s recent_step=1.47s eta=03:32:05 +2026-08-26 13:33:49 | INFO | train rank=0 | step=90500/100000 progress=90.5% loss=0.047585 lr=2.447174e-07 elapsed=33:38:19 avg_step=1.34s recent_step=1.62s eta=03:31:52 +2026-08-26 13:34:05 | INFO | train rank=0 | step=90510/100000 progress=90.5% loss=0.068924 lr=2.442067e-07 elapsed=33:38:35 avg_step=1.34s recent_step=1.61s eta=03:31:38 +2026-08-26 13:34:16 | INFO | train rank=0 | step=90520/100000 progress=90.5% loss=0.020515 lr=2.436966e-07 elapsed=33:38:46 avg_step=1.34s recent_step=1.16s eta=03:31:25 +2026-08-26 13:34:28 | INFO | train rank=0 | step=90530/100000 progress=90.5% loss=0.042618 lr=2.431869e-07 elapsed=33:38:58 avg_step=1.34s recent_step=1.19s eta=03:31:11 +2026-08-26 13:34:44 | INFO | train rank=0 | step=90540/100000 progress=90.5% loss=0.055786 lr=2.426778e-07 elapsed=33:39:14 avg_step=1.34s recent_step=1.55s eta=03:30:58 +2026-08-26 13:34:56 | INFO | train rank=0 | step=90550/100000 progress=90.5% loss=0.035622 lr=2.421692e-07 elapsed=33:39:26 avg_step=1.34s recent_step=1.20s eta=03:30:45 +2026-08-26 13:35:08 | INFO | train rank=0 | step=90560/100000 progress=90.6% loss=0.069275 lr=2.416611e-07 elapsed=33:39:38 avg_step=1.34s recent_step=1.22s eta=03:30:31 +2026-08-26 13:35:20 | INFO | train rank=0 | step=90570/100000 progress=90.6% loss=0.033539 lr=2.411535e-07 elapsed=33:39:50 avg_step=1.34s recent_step=1.22s eta=03:30:18 +2026-08-26 13:35:36 | INFO | train rank=0 | step=90580/100000 progress=90.6% loss=0.040329 lr=2.406465e-07 elapsed=33:40:06 avg_step=1.34s recent_step=1.57s eta=03:30:05 +2026-08-26 13:35:47 | INFO | train rank=0 | step=90590/100000 progress=90.6% loss=0.039810 lr=2.401399e-07 elapsed=33:40:17 avg_step=1.34s recent_step=1.16s eta=03:29:51 +2026-08-26 13:36:00 | INFO | train rank=0 | step=90600/100000 progress=90.6% loss=0.072052 lr=2.396339e-07 elapsed=33:40:29 avg_step=1.34s recent_step=1.22s eta=03:29:37 +2026-08-26 13:36:21 | INFO | train rank=0 | step=90610/100000 progress=90.6% loss=0.041199 lr=2.391284e-07 elapsed=33:40:51 avg_step=1.34s recent_step=2.19s eta=03:29:25 +2026-08-26 13:36:36 | INFO | train rank=0 | step=90620/100000 progress=90.6% loss=0.045700 lr=2.386235e-07 elapsed=33:41:06 avg_step=1.34s recent_step=1.46s eta=03:29:12 +2026-08-26 13:36:50 | INFO | train rank=0 | step=90630/100000 progress=90.6% loss=0.037880 lr=2.381190e-07 elapsed=33:41:20 avg_step=1.34s recent_step=1.39s eta=03:28:58 +2026-08-26 13:37:02 | INFO | train rank=0 | step=90640/100000 progress=90.6% loss=0.060913 lr=2.376151e-07 elapsed=33:41:32 avg_step=1.34s recent_step=1.17s eta=03:28:45 +2026-08-26 13:37:14 | INFO | train rank=0 | step=90650/100000 progress=90.6% loss=0.029381 lr=2.371117e-07 elapsed=33:41:43 avg_step=1.34s recent_step=1.19s eta=03:28:31 +2026-08-26 13:37:27 | INFO | train rank=0 | step=90660/100000 progress=90.7% loss=0.038925 lr=2.366088e-07 elapsed=33:41:56 avg_step=1.34s recent_step=1.30s eta=03:28:18 +2026-08-26 13:37:42 | INFO | train rank=0 | step=90670/100000 progress=90.7% loss=0.058044 lr=2.361065e-07 elapsed=33:42:11 avg_step=1.34s recent_step=1.50s eta=03:28:05 +2026-08-26 13:37:53 | INFO | train rank=0 | step=90680/100000 progress=90.7% loss=0.045181 lr=2.356046e-07 elapsed=33:42:23 avg_step=1.34s recent_step=1.18s eta=03:27:51 +2026-08-26 13:38:10 | INFO | train rank=0 | step=90690/100000 progress=90.7% loss=0.041580 lr=2.351033e-07 elapsed=33:42:40 avg_step=1.34s recent_step=1.66s eta=03:27:38 +2026-08-26 13:38:22 | INFO | train rank=0 | step=90700/100000 progress=90.7% loss=0.037872 lr=2.346025e-07 elapsed=33:42:51 avg_step=1.34s recent_step=1.16s eta=03:27:24 +2026-08-26 13:38:37 | INFO | train rank=0 | step=90710/100000 progress=90.7% loss=0.044006 lr=2.341022e-07 elapsed=33:43:07 avg_step=1.34s recent_step=1.56s eta=03:27:11 +2026-08-26 13:38:49 | INFO | train rank=0 | step=90720/100000 progress=90.7% loss=0.050751 lr=2.336025e-07 elapsed=33:43:19 avg_step=1.34s recent_step=1.16s eta=03:26:58 +2026-08-26 13:39:05 | INFO | train rank=0 | step=90730/100000 progress=90.7% loss=0.057587 lr=2.331032e-07 elapsed=33:43:35 avg_step=1.34s recent_step=1.60s eta=03:26:45 +2026-08-26 13:39:16 | INFO | train rank=0 | step=90740/100000 progress=90.7% loss=0.016573 lr=2.326045e-07 elapsed=33:43:46 avg_step=1.34s recent_step=1.13s eta=03:26:31 +2026-08-26 13:39:28 | INFO | train rank=0 | step=90750/100000 progress=90.8% loss=0.041946 lr=2.321063e-07 elapsed=33:43:58 avg_step=1.34s recent_step=1.19s eta=03:26:18 +2026-08-26 13:39:40 | INFO | train rank=0 | step=90760/100000 progress=90.8% loss=0.032242 lr=2.316086e-07 elapsed=33:44:10 avg_step=1.34s recent_step=1.21s eta=03:26:04 +2026-08-26 13:39:55 | INFO | train rank=0 | step=90770/100000 progress=90.8% loss=0.027992 lr=2.311115e-07 elapsed=33:44:25 avg_step=1.34s recent_step=1.49s eta=03:25:51 +2026-08-26 13:40:07 | INFO | train rank=0 | step=90780/100000 progress=90.8% loss=0.038635 lr=2.306149e-07 elapsed=33:44:37 avg_step=1.34s recent_step=1.22s eta=03:25:37 +2026-08-26 13:40:19 | INFO | train rank=0 | step=90790/100000 progress=90.8% loss=0.051575 lr=2.301188e-07 elapsed=33:44:49 avg_step=1.34s recent_step=1.18s eta=03:25:24 +2026-08-26 13:40:32 | INFO | train rank=0 | step=90800/100000 progress=90.8% loss=0.061096 lr=2.296232e-07 elapsed=33:45:02 avg_step=1.34s recent_step=1.32s eta=03:25:10 +2026-08-26 13:40:45 | INFO | train rank=0 | step=90810/100000 progress=90.8% loss=0.033424 lr=2.291281e-07 elapsed=33:45:15 avg_step=1.34s recent_step=1.31s eta=03:24:57 +2026-08-26 13:40:58 | INFO | train rank=0 | step=90820/100000 progress=90.8% loss=0.064163 lr=2.286336e-07 elapsed=33:45:28 avg_step=1.34s recent_step=1.30s eta=03:24:43 +2026-08-26 13:41:15 | INFO | train rank=0 | step=90830/100000 progress=90.8% loss=0.035477 lr=2.281395e-07 elapsed=33:45:44 avg_step=1.34s recent_step=1.65s eta=03:24:30 +2026-08-26 13:41:26 | INFO | train rank=0 | step=90840/100000 progress=90.8% loss=0.041351 lr=2.276460e-07 elapsed=33:45:56 avg_step=1.34s recent_step=1.19s eta=03:24:17 +2026-08-26 13:41:45 | INFO | train rank=0 | step=90850/100000 progress=90.8% loss=0.023422 lr=2.271531e-07 elapsed=33:46:15 avg_step=1.34s recent_step=1.88s eta=03:24:04 +2026-08-26 13:41:57 | INFO | train rank=0 | step=90860/100000 progress=90.9% loss=0.024933 lr=2.266606e-07 elapsed=33:46:27 avg_step=1.34s recent_step=1.20s eta=03:23:51 +2026-08-26 13:42:12 | INFO | train rank=0 | step=90870/100000 progress=90.9% loss=0.052155 lr=2.261687e-07 elapsed=33:46:42 avg_step=1.34s recent_step=1.50s eta=03:23:37 +2026-08-26 13:42:23 | INFO | train rank=0 | step=90880/100000 progress=90.9% loss=0.025143 lr=2.256773e-07 elapsed=33:46:53 avg_step=1.34s recent_step=1.10s eta=03:23:24 +2026-08-26 13:42:35 | INFO | train rank=0 | step=90890/100000 progress=90.9% loss=0.032742 lr=2.251864e-07 elapsed=33:47:05 avg_step=1.34s recent_step=1.19s eta=03:23:10 +2026-08-26 13:42:48 | INFO | train rank=0 | step=90900/100000 progress=90.9% loss=0.035442 lr=2.246960e-07 elapsed=33:47:18 avg_step=1.34s recent_step=1.25s eta=03:22:57 +2026-08-26 13:42:59 | INFO | train rank=0 | step=90910/100000 progress=90.9% loss=0.038635 lr=2.242062e-07 elapsed=33:47:29 avg_step=1.34s recent_step=1.10s eta=03:22:43 +2026-08-26 13:43:14 | INFO | train rank=0 | step=90920/100000 progress=90.9% loss=0.035500 lr=2.237169e-07 elapsed=33:47:44 avg_step=1.34s recent_step=1.54s eta=03:22:30 +2026-08-26 13:43:25 | INFO | train rank=0 | step=90930/100000 progress=90.9% loss=0.036209 lr=2.232281e-07 elapsed=33:47:55 avg_step=1.34s recent_step=1.11s eta=03:22:16 +2026-08-26 13:43:39 | INFO | train rank=0 | step=90940/100000 progress=90.9% loss=0.048828 lr=2.227398e-07 elapsed=33:48:09 avg_step=1.34s recent_step=1.35s eta=03:22:03 +2026-08-26 13:43:54 | INFO | train rank=0 | step=90950/100000 progress=91.0% loss=0.040985 lr=2.222520e-07 elapsed=33:48:24 avg_step=1.34s recent_step=1.49s eta=03:21:50 +2026-08-26 13:44:05 | INFO | train rank=0 | step=90960/100000 progress=91.0% loss=0.064484 lr=2.217648e-07 elapsed=33:48:35 avg_step=1.34s recent_step=1.12s eta=03:21:36 +2026-08-26 13:44:22 | INFO | train rank=0 | step=90970/100000 progress=91.0% loss=0.050323 lr=2.212781e-07 elapsed=33:48:52 avg_step=1.34s recent_step=1.74s eta=03:21:23 +2026-08-26 13:44:34 | INFO | train rank=0 | step=90980/100000 progress=91.0% loss=0.017403 lr=2.207919e-07 elapsed=33:49:03 avg_step=1.34s recent_step=1.12s eta=03:21:10 +2026-08-26 13:44:48 | INFO | train rank=0 | step=90990/100000 progress=91.0% loss=0.050323 lr=2.203062e-07 elapsed=33:49:18 avg_step=1.34s recent_step=1.42s eta=03:20:56 +2026-08-26 13:45:00 | INFO | train rank=0 | step=91000/100000 progress=91.0% loss=0.080444 lr=2.198211e-07 elapsed=33:49:30 avg_step=1.34s recent_step=1.21s eta=03:20:43 +2026-08-26 13:45:11 | INFO | train rank=0 | step=91010/100000 progress=91.0% loss=0.057068 lr=2.193365e-07 elapsed=33:49:41 avg_step=1.34s recent_step=1.11s eta=03:20:29 +2026-08-26 13:45:22 | INFO | train rank=0 | step=91020/100000 progress=91.0% loss=0.034828 lr=2.188524e-07 elapsed=33:49:52 avg_step=1.34s recent_step=1.14s eta=03:20:16 +2026-08-26 13:45:38 | INFO | train rank=0 | step=91030/100000 progress=91.0% loss=0.045326 lr=2.183688e-07 elapsed=33:50:08 avg_step=1.34s recent_step=1.53s eta=03:20:02 +2026-08-26 13:45:50 | INFO | train rank=0 | step=91040/100000 progress=91.0% loss=0.017895 lr=2.178858e-07 elapsed=33:50:20 avg_step=1.34s recent_step=1.22s eta=03:19:49 +2026-08-26 13:46:06 | INFO | train rank=0 | step=91050/100000 progress=91.0% loss=0.051743 lr=2.174032e-07 elapsed=33:50:35 avg_step=1.34s recent_step=1.57s eta=03:19:36 +2026-08-26 13:46:17 | INFO | train rank=0 | step=91060/100000 progress=91.1% loss=0.045120 lr=2.169212e-07 elapsed=33:50:47 avg_step=1.34s recent_step=1.18s eta=03:19:22 +2026-08-26 13:46:29 | INFO | train rank=0 | step=91070/100000 progress=91.1% loss=0.044449 lr=2.164397e-07 elapsed=33:50:59 avg_step=1.34s recent_step=1.16s eta=03:19:09 +2026-08-26 13:46:42 | INFO | train rank=0 | step=91080/100000 progress=91.1% loss=0.033676 lr=2.159588e-07 elapsed=33:51:12 avg_step=1.34s recent_step=1.35s eta=03:18:55 +2026-08-26 13:46:55 | INFO | train rank=0 | step=91090/100000 progress=91.1% loss=0.041855 lr=2.154783e-07 elapsed=33:51:25 avg_step=1.34s recent_step=1.28s eta=03:18:42 +2026-08-26 13:47:09 | INFO | train rank=0 | step=91100/100000 progress=91.1% loss=0.070068 lr=2.149984e-07 elapsed=33:51:39 avg_step=1.34s recent_step=1.37s eta=03:18:28 +2026-08-26 13:47:24 | INFO | train rank=0 | step=91110/100000 progress=91.1% loss=0.045273 lr=2.145190e-07 elapsed=33:51:54 avg_step=1.34s recent_step=1.54s eta=03:18:15 +2026-08-26 13:47:36 | INFO | train rank=0 | step=91120/100000 progress=91.1% loss=0.043671 lr=2.140402e-07 elapsed=33:52:06 avg_step=1.34s recent_step=1.16s eta=03:18:02 +2026-08-26 13:47:52 | INFO | train rank=0 | step=91130/100000 progress=91.1% loss=0.057190 lr=2.135618e-07 elapsed=33:52:22 avg_step=1.34s recent_step=1.60s eta=03:17:49 +2026-08-26 13:48:03 | INFO | train rank=0 | step=91140/100000 progress=91.1% loss=0.055511 lr=2.130840e-07 elapsed=33:52:33 avg_step=1.34s recent_step=1.15s eta=03:17:35 +2026-08-26 13:48:17 | INFO | train rank=0 | step=91150/100000 progress=91.1% loss=0.037361 lr=2.126067e-07 elapsed=33:52:47 avg_step=1.34s recent_step=1.34s eta=03:17:22 +2026-08-26 13:48:33 | INFO | train rank=0 | step=91160/100000 progress=91.2% loss=0.030602 lr=2.121300e-07 elapsed=33:53:03 avg_step=1.34s recent_step=1.61s eta=03:17:09 +2026-08-26 13:48:44 | INFO | train rank=0 | step=91170/100000 progress=91.2% loss=0.055298 lr=2.116537e-07 elapsed=33:53:14 avg_step=1.34s recent_step=1.13s eta=03:16:55 +2026-08-26 13:48:56 | INFO | train rank=0 | step=91180/100000 progress=91.2% loss=0.077606 lr=2.111780e-07 elapsed=33:53:26 avg_step=1.34s recent_step=1.22s eta=03:16:41 +2026-08-26 13:49:11 | INFO | train rank=0 | step=91190/100000 progress=91.2% loss=0.039276 lr=2.107028e-07 elapsed=33:53:41 avg_step=1.34s recent_step=1.42s eta=03:16:28 +2026-08-26 13:49:24 | INFO | train rank=0 | step=91200/100000 progress=91.2% loss=0.083923 lr=2.102281e-07 elapsed=33:53:54 avg_step=1.34s recent_step=1.34s eta=03:16:15 +2026-08-26 13:49:39 | INFO | train rank=0 | step=91210/100000 progress=91.2% loss=0.103333 lr=2.097540e-07 elapsed=33:54:09 avg_step=1.34s recent_step=1.53s eta=03:16:02 +2026-08-26 13:49:51 | INFO | train rank=0 | step=91220/100000 progress=91.2% loss=0.039326 lr=2.092803e-07 elapsed=33:54:21 avg_step=1.34s recent_step=1.20s eta=03:15:48 +2026-08-26 13:50:03 | INFO | train rank=0 | step=91230/100000 progress=91.2% loss=0.031231 lr=2.088072e-07 elapsed=33:54:33 avg_step=1.34s recent_step=1.19s eta=03:15:35 +2026-08-26 13:50:18 | INFO | train rank=0 | step=91240/100000 progress=91.2% loss=0.048294 lr=2.083346e-07 elapsed=33:54:48 avg_step=1.34s recent_step=1.50s eta=03:15:21 +2026-08-26 13:50:34 | INFO | train rank=0 | step=91250/100000 progress=91.2% loss=0.056335 lr=2.078626e-07 elapsed=33:55:03 avg_step=1.34s recent_step=1.54s eta=03:15:08 +2026-08-26 13:50:48 | INFO | train rank=0 | step=91260/100000 progress=91.3% loss=0.049301 lr=2.073911e-07 elapsed=33:55:18 avg_step=1.34s recent_step=1.42s eta=03:14:55 +2026-08-26 13:50:59 | INFO | train rank=0 | step=91270/100000 progress=91.3% loss=0.054588 lr=2.069200e-07 elapsed=33:55:29 avg_step=1.34s recent_step=1.15s eta=03:14:41 +2026-08-26 13:51:14 | INFO | train rank=0 | step=91280/100000 progress=91.3% loss=0.084930 lr=2.064496e-07 elapsed=33:55:44 avg_step=1.34s recent_step=1.51s eta=03:14:28 +2026-08-26 13:51:32 | INFO | train rank=0 | step=91290/100000 progress=91.3% loss=0.054443 lr=2.059796e-07 elapsed=33:56:02 avg_step=1.34s recent_step=1.75s eta=03:14:15 +2026-08-26 13:51:43 | INFO | train rank=0 | step=91300/100000 progress=91.3% loss=0.043289 lr=2.055102e-07 elapsed=33:56:13 avg_step=1.34s recent_step=1.14s eta=03:14:01 +2026-08-26 13:51:55 | INFO | train rank=0 | step=91310/100000 progress=91.3% loss=0.041183 lr=2.050413e-07 elapsed=33:56:25 avg_step=1.34s recent_step=1.17s eta=03:13:48 +2026-08-26 13:52:08 | INFO | train rank=0 | step=91320/100000 progress=91.3% loss=0.036453 lr=2.045729e-07 elapsed=33:56:38 avg_step=1.34s recent_step=1.34s eta=03:13:35 +2026-08-26 13:52:21 | INFO | train rank=0 | step=91330/100000 progress=91.3% loss=0.044968 lr=2.041050e-07 elapsed=33:56:51 avg_step=1.34s recent_step=1.26s eta=03:13:21 +2026-08-26 13:52:41 | INFO | train rank=0 | step=91340/100000 progress=91.3% loss=0.039597 lr=2.036377e-07 elapsed=33:57:11 avg_step=1.34s recent_step=2.00s eta=03:13:08 +2026-08-26 13:52:54 | INFO | train rank=0 | step=91350/100000 progress=91.3% loss=0.043259 lr=2.031709e-07 elapsed=33:57:24 avg_step=1.34s recent_step=1.27s eta=03:12:55 +2026-08-26 13:53:09 | INFO | train rank=0 | step=91360/100000 progress=91.4% loss=0.031631 lr=2.027046e-07 elapsed=33:57:39 avg_step=1.34s recent_step=1.56s eta=03:12:42 +2026-08-26 13:53:21 | INFO | train rank=0 | step=91370/100000 progress=91.4% loss=0.048264 lr=2.022388e-07 elapsed=33:57:51 avg_step=1.34s recent_step=1.14s eta=03:12:28 +2026-08-26 13:53:32 | INFO | train rank=0 | step=91380/100000 progress=91.4% loss=0.051987 lr=2.017736e-07 elapsed=33:58:02 avg_step=1.34s recent_step=1.16s eta=03:12:15 +2026-08-26 13:53:44 | INFO | train rank=0 | step=91390/100000 progress=91.4% loss=0.022865 lr=2.013088e-07 elapsed=33:58:14 avg_step=1.34s recent_step=1.14s eta=03:12:01 +2026-08-26 13:53:55 | INFO | train rank=0 | step=91400/100000 progress=91.4% loss=0.052094 lr=2.008447e-07 elapsed=33:58:25 avg_step=1.34s recent_step=1.15s eta=03:11:47 +2026-08-26 13:54:06 | INFO | train rank=0 | step=91410/100000 progress=91.4% loss=0.046204 lr=2.003810e-07 elapsed=33:58:36 avg_step=1.34s recent_step=1.09s eta=03:11:34 +2026-08-26 13:54:21 | INFO | train rank=0 | step=91420/100000 progress=91.4% loss=0.045654 lr=1.999178e-07 elapsed=33:58:51 avg_step=1.34s recent_step=1.52s eta=03:11:21 +2026-08-26 13:54:35 | INFO | train rank=0 | step=91430/100000 progress=91.4% loss=0.045868 lr=1.994552e-07 elapsed=33:59:05 avg_step=1.34s recent_step=1.39s eta=03:11:07 +2026-08-26 13:54:51 | INFO | train rank=0 | step=91440/100000 progress=91.4% loss=0.025146 lr=1.989931e-07 elapsed=33:59:21 avg_step=1.34s recent_step=1.58s eta=03:10:54 +2026-08-26 13:55:03 | INFO | train rank=0 | step=91450/100000 progress=91.5% loss=0.033081 lr=1.985316e-07 elapsed=33:59:33 avg_step=1.34s recent_step=1.23s eta=03:10:41 +2026-08-26 13:55:19 | INFO | train rank=0 | step=91460/100000 progress=91.5% loss=0.042526 lr=1.980705e-07 elapsed=33:59:49 avg_step=1.34s recent_step=1.56s eta=03:10:27 +2026-08-26 13:55:30 | INFO | train rank=0 | step=91470/100000 progress=91.5% loss=0.037743 lr=1.976100e-07 elapsed=34:00:00 avg_step=1.34s recent_step=1.15s eta=03:10:14 +2026-08-26 13:55:45 | INFO | train rank=0 | step=91480/100000 progress=91.5% loss=0.058136 lr=1.971500e-07 elapsed=34:00:15 avg_step=1.34s recent_step=1.46s eta=03:10:01 +2026-08-26 13:55:58 | INFO | train rank=0 | step=91490/100000 progress=91.5% loss=0.035080 lr=1.966906e-07 elapsed=34:00:27 avg_step=1.34s recent_step=1.26s eta=03:09:47 +2026-08-26 13:56:14 | INFO | train rank=0 | step=91500/100000 progress=91.5% loss=0.055115 lr=1.962316e-07 elapsed=34:00:44 avg_step=1.34s recent_step=1.67s eta=03:09:34 +2026-08-26 13:56:26 | INFO | train rank=0 | step=91510/100000 progress=91.5% loss=0.054352 lr=1.957732e-07 elapsed=34:00:56 avg_step=1.34s recent_step=1.21s eta=03:09:21 +2026-08-26 13:56:38 | INFO | train rank=0 | step=91520/100000 progress=91.5% loss=0.023445 lr=1.953153e-07 elapsed=34:01:08 avg_step=1.34s recent_step=1.17s eta=03:09:07 +2026-08-26 13:56:51 | INFO | train rank=0 | step=91530/100000 progress=91.5% loss=0.044586 lr=1.948580e-07 elapsed=34:01:21 avg_step=1.34s recent_step=1.33s eta=03:08:54 +2026-08-26 13:57:04 | INFO | train rank=0 | step=91540/100000 progress=91.5% loss=0.054466 lr=1.944011e-07 elapsed=34:01:33 avg_step=1.34s recent_step=1.22s eta=03:08:40 +2026-08-26 13:57:19 | INFO | train rank=0 | step=91550/100000 progress=91.5% loss=0.073990 lr=1.939448e-07 elapsed=34:01:49 avg_step=1.34s recent_step=1.55s eta=03:08:27 +2026-08-26 13:57:31 | INFO | train rank=0 | step=91560/100000 progress=91.6% loss=0.076202 lr=1.934890e-07 elapsed=34:02:01 avg_step=1.34s recent_step=1.25s eta=03:08:14 +2026-08-26 13:57:43 | INFO | train rank=0 | step=91570/100000 progress=91.6% loss=0.040054 lr=1.930338e-07 elapsed=34:02:13 avg_step=1.34s recent_step=1.13s eta=03:08:00 +2026-08-26 13:57:55 | INFO | train rank=0 | step=91580/100000 progress=91.6% loss=0.055298 lr=1.925790e-07 elapsed=34:02:25 avg_step=1.34s recent_step=1.24s eta=03:07:47 +2026-08-26 13:58:10 | INFO | train rank=0 | step=91590/100000 progress=91.6% loss=0.025803 lr=1.921248e-07 elapsed=34:02:40 avg_step=1.34s recent_step=1.48s eta=03:07:33 +2026-08-26 13:58:24 | INFO | train rank=0 | step=91600/100000 progress=91.6% loss=0.042625 lr=1.916711e-07 elapsed=34:02:54 avg_step=1.34s recent_step=1.37s eta=03:07:20 +2026-08-26 13:58:37 | INFO | train rank=0 | step=91610/100000 progress=91.6% loss=0.052063 lr=1.912180e-07 elapsed=34:03:07 avg_step=1.34s recent_step=1.37s eta=03:07:07 +2026-08-26 13:58:49 | INFO | train rank=0 | step=91620/100000 progress=91.6% loss=0.072021 lr=1.907653e-07 elapsed=34:03:19 avg_step=1.34s recent_step=1.14s eta=03:06:53 +2026-08-26 13:59:00 | INFO | train rank=0 | step=91630/100000 progress=91.6% loss=0.021614 lr=1.903132e-07 elapsed=34:03:30 avg_step=1.34s recent_step=1.15s eta=03:06:39 +2026-08-26 13:59:12 | INFO | train rank=0 | step=91640/100000 progress=91.6% loss=0.053009 lr=1.898616e-07 elapsed=34:03:42 avg_step=1.34s recent_step=1.18s eta=03:06:26 +2026-08-26 13:59:27 | INFO | train rank=0 | step=91650/100000 progress=91.6% loss=0.053314 lr=1.894106e-07 elapsed=34:03:57 avg_step=1.34s recent_step=1.48s eta=03:06:13 +2026-08-26 13:59:40 | INFO | train rank=0 | step=91660/100000 progress=91.7% loss=0.038185 lr=1.889601e-07 elapsed=34:04:10 avg_step=1.34s recent_step=1.34s eta=03:05:59 +2026-08-26 13:59:54 | INFO | train rank=0 | step=91670/100000 progress=91.7% loss=0.048691 lr=1.885101e-07 elapsed=34:04:24 avg_step=1.34s recent_step=1.39s eta=03:05:46 +2026-08-26 14:00:07 | INFO | train rank=0 | step=91680/100000 progress=91.7% loss=0.038925 lr=1.880606e-07 elapsed=34:04:37 avg_step=1.34s recent_step=1.26s eta=03:05:33 +2026-08-26 14:00:22 | INFO | train rank=0 | step=91690/100000 progress=91.7% loss=0.039421 lr=1.876116e-07 elapsed=34:04:52 avg_step=1.34s recent_step=1.50s eta=03:05:19 +2026-08-26 14:00:34 | INFO | train rank=0 | step=91700/100000 progress=91.7% loss=0.030533 lr=1.871632e-07 elapsed=34:05:04 avg_step=1.34s recent_step=1.24s eta=03:05:06 +2026-08-26 14:00:48 | INFO | train rank=0 | step=91710/100000 progress=91.7% loss=0.053436 lr=1.867153e-07 elapsed=34:05:18 avg_step=1.34s recent_step=1.41s eta=03:04:52 +2026-08-26 14:01:00 | INFO | train rank=0 | step=91720/100000 progress=91.7% loss=0.028656 lr=1.862679e-07 elapsed=34:05:30 avg_step=1.34s recent_step=1.21s eta=03:04:39 +2026-08-26 14:01:12 | INFO | train rank=0 | step=91730/100000 progress=91.7% loss=0.044189 lr=1.858211e-07 elapsed=34:05:42 avg_step=1.34s recent_step=1.20s eta=03:04:25 +2026-08-26 14:01:29 | INFO | train rank=0 | step=91740/100000 progress=91.7% loss=0.048553 lr=1.853748e-07 elapsed=34:05:59 avg_step=1.34s recent_step=1.71s eta=03:04:12 +2026-08-26 14:01:41 | INFO | train rank=0 | step=91750/100000 progress=91.8% loss=0.033539 lr=1.849290e-07 elapsed=34:06:11 avg_step=1.34s recent_step=1.15s eta=03:03:59 +2026-08-26 14:01:55 | INFO | train rank=0 | step=91760/100000 progress=91.8% loss=0.024765 lr=1.844837e-07 elapsed=34:06:25 avg_step=1.34s recent_step=1.37s eta=03:03:46 +2026-08-26 14:02:08 | INFO | train rank=0 | step=91770/100000 progress=91.8% loss=0.040558 lr=1.840390e-07 elapsed=34:06:38 avg_step=1.34s recent_step=1.37s eta=03:03:32 +2026-08-26 14:02:20 | INFO | train rank=0 | step=91780/100000 progress=91.8% loss=0.042801 lr=1.835948e-07 elapsed=34:06:50 avg_step=1.34s recent_step=1.17s eta=03:03:19 +2026-08-26 14:02:35 | INFO | train rank=0 | step=91790/100000 progress=91.8% loss=0.031441 lr=1.831511e-07 elapsed=34:07:05 avg_step=1.34s recent_step=1.53s eta=03:03:05 +2026-08-26 14:02:51 | INFO | train rank=0 | step=91800/100000 progress=91.8% loss=0.045258 lr=1.827079e-07 elapsed=34:07:21 avg_step=1.34s recent_step=1.60s eta=03:02:52 +2026-08-26 14:03:06 | INFO | train rank=0 | step=91810/100000 progress=91.8% loss=0.068115 lr=1.822653e-07 elapsed=34:07:35 avg_step=1.34s recent_step=1.43s eta=03:02:39 +2026-08-26 14:03:21 | INFO | train rank=0 | step=91820/100000 progress=91.8% loss=0.033722 lr=1.818232e-07 elapsed=34:07:50 avg_step=1.34s recent_step=1.50s eta=03:02:26 +2026-08-26 14:03:32 | INFO | train rank=0 | step=91830/100000 progress=91.8% loss=0.059341 lr=1.813816e-07 elapsed=34:08:02 avg_step=1.34s recent_step=1.17s eta=03:02:12 +2026-08-26 14:03:46 | INFO | train rank=0 | step=91840/100000 progress=91.8% loss=0.038704 lr=1.809406e-07 elapsed=34:08:16 avg_step=1.34s recent_step=1.39s eta=03:01:59 +2026-08-26 14:04:00 | INFO | train rank=0 | step=91850/100000 progress=91.8% loss=0.025070 lr=1.805000e-07 elapsed=34:08:30 avg_step=1.34s recent_step=1.39s eta=03:01:46 +2026-08-26 14:04:13 | INFO | train rank=0 | step=91860/100000 progress=91.9% loss=0.040176 lr=1.800600e-07 elapsed=34:08:43 avg_step=1.34s recent_step=1.30s eta=03:01:32 +2026-08-26 14:04:26 | INFO | train rank=0 | step=91870/100000 progress=91.9% loss=0.038574 lr=1.796206e-07 elapsed=34:08:55 avg_step=1.34s recent_step=1.25s eta=03:01:19 +2026-08-26 14:04:40 | INFO | train rank=0 | step=91880/100000 progress=91.9% loss=0.028343 lr=1.791816e-07 elapsed=34:09:10 avg_step=1.34s recent_step=1.48s eta=03:01:05 +2026-08-26 14:04:55 | INFO | train rank=0 | step=91890/100000 progress=91.9% loss=0.035454 lr=1.787432e-07 elapsed=34:09:25 avg_step=1.34s recent_step=1.43s eta=03:00:52 +2026-08-26 14:05:10 | INFO | train rank=0 | step=91900/100000 progress=91.9% loss=0.033890 lr=1.783053e-07 elapsed=34:09:40 avg_step=1.34s recent_step=1.50s eta=03:00:39 +2026-08-26 14:05:25 | INFO | train rank=0 | step=91910/100000 progress=91.9% loss=0.048813 lr=1.778680e-07 elapsed=34:09:55 avg_step=1.34s recent_step=1.57s eta=03:00:26 +2026-08-26 14:05:40 | INFO | train rank=0 | step=91920/100000 progress=91.9% loss=0.041199 lr=1.774311e-07 elapsed=34:10:10 avg_step=1.34s recent_step=1.46s eta=03:00:12 +2026-08-26 14:05:51 | INFO | train rank=0 | step=91930/100000 progress=91.9% loss=0.031204 lr=1.769948e-07 elapsed=34:10:21 avg_step=1.34s recent_step=1.11s eta=02:59:59 +2026-08-26 14:06:07 | INFO | train rank=0 | step=91940/100000 progress=91.9% loss=0.043472 lr=1.765590e-07 elapsed=34:10:37 avg_step=1.34s recent_step=1.56s eta=02:59:46 +2026-08-26 14:06:20 | INFO | train rank=0 | step=91950/100000 progress=92.0% loss=0.049881 lr=1.761238e-07 elapsed=34:10:50 avg_step=1.34s recent_step=1.37s eta=02:59:32 +2026-08-26 14:06:33 | INFO | train rank=0 | step=91960/100000 progress=92.0% loss=0.040997 lr=1.756891e-07 elapsed=34:11:03 avg_step=1.34s recent_step=1.23s eta=02:59:19 +2026-08-26 14:06:48 | INFO | train rank=0 | step=91970/100000 progress=92.0% loss=0.035172 lr=1.752549e-07 elapsed=34:11:17 avg_step=1.34s recent_step=1.48s eta=02:59:06 +2026-08-26 14:07:03 | INFO | train rank=0 | step=91980/100000 progress=92.0% loss=0.050537 lr=1.748212e-07 elapsed=34:11:32 avg_step=1.34s recent_step=1.51s eta=02:58:52 +2026-08-26 14:07:15 | INFO | train rank=0 | step=91990/100000 progress=92.0% loss=0.069427 lr=1.743881e-07 elapsed=34:11:45 avg_step=1.34s recent_step=1.22s eta=02:58:39 +2026-08-26 14:07:29 | INFO | train rank=0 | step=92000/100000 progress=92.0% loss=0.049530 lr=1.739554e-07 elapsed=34:11:59 avg_step=1.34s recent_step=1.41s eta=02:58:26 +2026-08-26 14:07:41 | INFO | train rank=0 | step=92010/100000 progress=92.0% loss=0.051971 lr=1.735234e-07 elapsed=34:12:11 avg_step=1.34s recent_step=1.18s eta=02:58:12 +2026-08-26 14:07:58 | INFO | train rank=0 | step=92020/100000 progress=92.0% loss=0.031578 lr=1.730918e-07 elapsed=34:12:28 avg_step=1.34s recent_step=1.78s eta=02:57:59 +2026-08-26 14:08:10 | INFO | train rank=0 | step=92030/100000 progress=92.0% loss=0.043259 lr=1.726608e-07 elapsed=34:12:40 avg_step=1.34s recent_step=1.18s eta=02:57:45 +2026-08-26 14:08:22 | INFO | train rank=0 | step=92040/100000 progress=92.0% loss=0.057770 lr=1.722303e-07 elapsed=34:12:52 avg_step=1.34s recent_step=1.21s eta=02:57:32 +2026-08-26 14:08:34 | INFO | train rank=0 | step=92050/100000 progress=92.0% loss=0.028580 lr=1.718003e-07 elapsed=34:13:04 avg_step=1.34s recent_step=1.18s eta=02:57:18 +2026-08-26 14:08:46 | INFO | train rank=0 | step=92060/100000 progress=92.1% loss=0.043976 lr=1.713709e-07 elapsed=34:13:16 avg_step=1.34s recent_step=1.20s eta=02:57:05 +2026-08-26 14:08:58 | INFO | train rank=0 | step=92070/100000 progress=92.1% loss=0.029198 lr=1.709419e-07 elapsed=34:13:28 avg_step=1.34s recent_step=1.18s eta=02:56:51 +2026-08-26 14:09:10 | INFO | train rank=0 | step=92080/100000 progress=92.1% loss=0.043274 lr=1.705135e-07 elapsed=34:13:40 avg_step=1.34s recent_step=1.18s eta=02:56:38 +2026-08-26 14:09:22 | INFO | train rank=0 | step=92090/100000 progress=92.1% loss=0.046112 lr=1.700857e-07 elapsed=34:13:52 avg_step=1.34s recent_step=1.20s eta=02:56:24 +2026-08-26 14:09:33 | INFO | train rank=0 | step=92100/100000 progress=92.1% loss=0.040749 lr=1.696584e-07 elapsed=34:14:03 avg_step=1.34s recent_step=1.16s eta=02:56:11 +2026-08-26 14:09:50 | INFO | train rank=0 | step=92110/100000 progress=92.1% loss=0.032486 lr=1.692315e-07 elapsed=34:14:20 avg_step=1.34s recent_step=1.71s eta=02:55:58 +2026-08-26 14:10:06 | INFO | train rank=0 | step=92120/100000 progress=92.1% loss=0.066162 lr=1.688053e-07 elapsed=34:14:36 avg_step=1.34s recent_step=1.55s eta=02:55:45 +2026-08-26 14:10:20 | INFO | train rank=0 | step=92130/100000 progress=92.1% loss=0.029144 lr=1.683795e-07 elapsed=34:14:50 avg_step=1.34s recent_step=1.39s eta=02:55:31 +2026-08-26 14:10:33 | INFO | train rank=0 | step=92140/100000 progress=92.1% loss=0.052429 lr=1.679543e-07 elapsed=34:15:03 avg_step=1.34s recent_step=1.37s eta=02:55:18 +2026-08-26 14:10:48 | INFO | train rank=0 | step=92150/100000 progress=92.2% loss=0.036972 lr=1.675296e-07 elapsed=34:15:18 avg_step=1.34s recent_step=1.49s eta=02:55:05 +2026-08-26 14:11:00 | INFO | train rank=0 | step=92160/100000 progress=92.2% loss=0.078430 lr=1.671054e-07 elapsed=34:15:30 avg_step=1.34s recent_step=1.19s eta=02:54:51 +2026-08-26 14:11:13 | INFO | train rank=0 | step=92170/100000 progress=92.2% loss=0.041550 lr=1.666818e-07 elapsed=34:15:42 avg_step=1.34s recent_step=1.24s eta=02:54:38 +2026-08-26 14:11:28 | INFO | train rank=0 | step=92180/100000 progress=92.2% loss=0.051544 lr=1.662587e-07 elapsed=34:15:58 avg_step=1.34s recent_step=1.54s eta=02:54:24 +2026-08-26 14:11:40 | INFO | train rank=0 | step=92190/100000 progress=92.2% loss=0.034668 lr=1.658361e-07 elapsed=34:16:10 avg_step=1.34s recent_step=1.19s eta=02:54:11 +2026-08-26 14:11:54 | INFO | train rank=0 | step=92200/100000 progress=92.2% loss=0.082611 lr=1.654141e-07 elapsed=34:16:24 avg_step=1.34s recent_step=1.41s eta=02:53:58 +2026-08-26 14:12:05 | INFO | train rank=0 | step=92210/100000 progress=92.2% loss=0.061417 lr=1.649926e-07 elapsed=34:16:35 avg_step=1.34s recent_step=1.11s eta=02:53:44 +2026-08-26 14:12:19 | INFO | train rank=0 | step=92220/100000 progress=92.2% loss=0.039139 lr=1.645716e-07 elapsed=34:16:49 avg_step=1.34s recent_step=1.42s eta=02:53:31 +2026-08-26 14:12:33 | INFO | train rank=0 | step=92230/100000 progress=92.2% loss=0.054413 lr=1.641511e-07 elapsed=34:17:03 avg_step=1.34s recent_step=1.41s eta=02:53:17 +2026-08-26 14:12:45 | INFO | train rank=0 | step=92240/100000 progress=92.2% loss=0.037735 lr=1.637312e-07 elapsed=34:17:15 avg_step=1.34s recent_step=1.20s eta=02:53:04 +2026-08-26 14:12:57 | INFO | train rank=0 | step=92250/100000 progress=92.2% loss=0.046860 lr=1.633118e-07 elapsed=34:17:27 avg_step=1.34s recent_step=1.21s eta=02:52:50 +2026-08-26 14:13:13 | INFO | train rank=0 | step=92260/100000 progress=92.3% loss=0.043198 lr=1.628929e-07 elapsed=34:17:43 avg_step=1.34s recent_step=1.59s eta=02:52:37 +2026-08-26 14:13:27 | INFO | train rank=0 | step=92270/100000 progress=92.3% loss=0.042007 lr=1.624745e-07 elapsed=34:17:57 avg_step=1.34s recent_step=1.40s eta=02:52:24 +2026-08-26 14:13:43 | INFO | train rank=0 | step=92280/100000 progress=92.3% loss=0.048859 lr=1.620567e-07 elapsed=34:18:13 avg_step=1.34s recent_step=1.55s eta=02:52:11 +2026-08-26 14:13:54 | INFO | train rank=0 | step=92290/100000 progress=92.3% loss=0.069122 lr=1.616394e-07 elapsed=34:18:24 avg_step=1.34s recent_step=1.11s eta=02:51:57 +2026-08-26 14:14:09 | INFO | train rank=0 | step=92300/100000 progress=92.3% loss=0.047424 lr=1.612227e-07 elapsed=34:18:39 avg_step=1.34s recent_step=1.48s eta=02:51:44 +2026-08-26 14:14:22 | INFO | train rank=0 | step=92310/100000 progress=92.3% loss=0.036621 lr=1.608064e-07 elapsed=34:18:52 avg_step=1.34s recent_step=1.31s eta=02:51:31 +2026-08-26 14:14:38 | INFO | train rank=0 | step=92320/100000 progress=92.3% loss=0.046951 lr=1.603907e-07 elapsed=34:19:07 avg_step=1.34s recent_step=1.58s eta=02:51:17 +2026-08-26 14:14:50 | INFO | train rank=0 | step=92330/100000 progress=92.3% loss=0.048477 lr=1.599756e-07 elapsed=34:19:20 avg_step=1.34s recent_step=1.26s eta=02:51:04 +2026-08-26 14:15:05 | INFO | train rank=0 | step=92340/100000 progress=92.3% loss=0.054565 lr=1.595609e-07 elapsed=34:19:35 avg_step=1.34s recent_step=1.47s eta=02:50:51 +2026-08-26 14:15:17 | INFO | train rank=0 | step=92350/100000 progress=92.3% loss=0.047638 lr=1.591468e-07 elapsed=34:19:47 avg_step=1.34s recent_step=1.18s eta=02:50:37 +2026-08-26 14:15:29 | INFO | train rank=0 | step=92360/100000 progress=92.4% loss=0.066391 lr=1.587332e-07 elapsed=34:19:59 avg_step=1.34s recent_step=1.20s eta=02:50:24 +2026-08-26 14:15:43 | INFO | train rank=0 | step=92370/100000 progress=92.4% loss=0.044006 lr=1.583202e-07 elapsed=34:20:13 avg_step=1.34s recent_step=1.45s eta=02:50:10 +2026-08-26 14:15:58 | INFO | train rank=0 | step=92380/100000 progress=92.4% loss=0.054474 lr=1.579077e-07 elapsed=34:20:28 avg_step=1.34s recent_step=1.46s eta=02:49:57 +2026-08-26 14:16:09 | INFO | train rank=0 | step=92390/100000 progress=92.4% loss=0.045410 lr=1.574957e-07 elapsed=34:20:39 avg_step=1.34s recent_step=1.13s eta=02:49:43 +2026-08-26 14:16:26 | INFO | train rank=0 | step=92400/100000 progress=92.4% loss=0.050751 lr=1.570842e-07 elapsed=34:20:56 avg_step=1.34s recent_step=1.67s eta=02:49:30 +2026-08-26 14:16:39 | INFO | train rank=0 | step=92410/100000 progress=92.4% loss=0.048004 lr=1.566733e-07 elapsed=34:21:09 avg_step=1.34s recent_step=1.33s eta=02:49:17 +2026-08-26 14:16:51 | INFO | train rank=0 | step=92420/100000 progress=92.4% loss=0.040321 lr=1.562629e-07 elapsed=34:21:20 avg_step=1.34s recent_step=1.15s eta=02:49:03 +2026-08-26 14:17:03 | INFO | train rank=0 | step=92430/100000 progress=92.4% loss=0.049850 lr=1.558530e-07 elapsed=34:21:33 avg_step=1.34s recent_step=1.23s eta=02:48:50 +2026-08-26 14:17:18 | INFO | train rank=0 | step=92440/100000 progress=92.4% loss=0.058655 lr=1.554436e-07 elapsed=34:21:48 avg_step=1.34s recent_step=1.52s eta=02:48:37 +2026-08-26 14:17:32 | INFO | train rank=0 | step=92450/100000 progress=92.5% loss=0.063110 lr=1.550348e-07 elapsed=34:22:02 avg_step=1.34s recent_step=1.40s eta=02:48:23 +2026-08-26 14:17:47 | INFO | train rank=0 | step=92460/100000 progress=92.5% loss=0.029297 lr=1.546265e-07 elapsed=34:22:17 avg_step=1.34s recent_step=1.53s eta=02:48:10 +2026-08-26 14:17:59 | INFO | train rank=0 | step=92470/100000 progress=92.5% loss=0.053406 lr=1.542188e-07 elapsed=34:22:29 avg_step=1.34s recent_step=1.16s eta=02:47:57 +2026-08-26 14:18:11 | INFO | train rank=0 | step=92480/100000 progress=92.5% loss=0.043304 lr=1.538115e-07 elapsed=34:22:41 avg_step=1.34s recent_step=1.19s eta=02:47:43 +2026-08-26 14:18:22 | INFO | train rank=0 | step=92490/100000 progress=92.5% loss=0.078461 lr=1.534048e-07 elapsed=34:22:52 avg_step=1.34s recent_step=1.15s eta=02:47:30 +2026-08-26 14:18:36 | INFO | train rank=0 | step=92500/100000 progress=92.5% loss=0.040756 lr=1.529987e-07 elapsed=34:23:06 avg_step=1.34s recent_step=1.33s eta=02:47:16 +2026-08-26 14:18:49 | INFO | train rank=0 | step=92510/100000 progress=92.5% loss=0.064697 lr=1.525930e-07 elapsed=34:23:19 avg_step=1.34s recent_step=1.34s eta=02:47:03 +2026-08-26 14:19:04 | INFO | train rank=0 | step=92520/100000 progress=92.5% loss=0.058273 lr=1.521879e-07 elapsed=34:23:34 avg_step=1.34s recent_step=1.50s eta=02:46:50 +2026-08-26 14:19:19 | INFO | train rank=0 | step=92530/100000 progress=92.5% loss=0.039955 lr=1.517833e-07 elapsed=34:23:49 avg_step=1.34s recent_step=1.51s eta=02:46:36 +2026-08-26 14:19:31 | INFO | train rank=0 | step=92540/100000 progress=92.5% loss=0.034470 lr=1.513793e-07 elapsed=34:24:01 avg_step=1.34s recent_step=1.24s eta=02:46:23 +2026-08-26 14:19:46 | INFO | train rank=0 | step=92550/100000 progress=92.5% loss=0.045242 lr=1.509758e-07 elapsed=34:24:16 avg_step=1.34s recent_step=1.45s eta=02:46:10 +2026-08-26 14:19:58 | INFO | train rank=0 | step=92560/100000 progress=92.6% loss=0.021080 lr=1.505728e-07 elapsed=34:24:28 avg_step=1.34s recent_step=1.19s eta=02:45:56 +2026-08-26 14:20:13 | INFO | train rank=0 | step=92570/100000 progress=92.6% loss=0.043365 lr=1.501703e-07 elapsed=34:24:43 avg_step=1.34s recent_step=1.56s eta=02:45:43 +2026-08-26 14:20:25 | INFO | train rank=0 | step=92580/100000 progress=92.6% loss=0.107300 lr=1.497684e-07 elapsed=34:24:55 avg_step=1.34s recent_step=1.17s eta=02:45:29 +2026-08-26 14:20:38 | INFO | train rank=0 | step=92590/100000 progress=92.6% loss=0.041367 lr=1.493670e-07 elapsed=34:25:08 avg_step=1.34s recent_step=1.29s eta=02:45:16 +2026-08-26 14:20:50 | INFO | train rank=0 | step=92600/100000 progress=92.6% loss=0.031967 lr=1.489662e-07 elapsed=34:25:20 avg_step=1.34s recent_step=1.20s eta=02:45:02 +2026-08-26 14:21:02 | INFO | train rank=0 | step=92610/100000 progress=92.6% loss=0.048080 lr=1.485658e-07 elapsed=34:25:32 avg_step=1.34s recent_step=1.21s eta=02:44:49 +2026-08-26 14:21:20 | INFO | train rank=0 | step=92620/100000 progress=92.6% loss=0.080322 lr=1.481660e-07 elapsed=34:25:49 avg_step=1.34s recent_step=1.75s eta=02:44:36 +2026-08-26 14:21:35 | INFO | train rank=0 | step=92630/100000 progress=92.6% loss=0.027138 lr=1.477667e-07 elapsed=34:26:05 avg_step=1.34s recent_step=1.59s eta=02:44:23 +2026-08-26 14:21:47 | INFO | train rank=0 | step=92640/100000 progress=92.6% loss=0.043839 lr=1.473680e-07 elapsed=34:26:17 avg_step=1.34s recent_step=1.18s eta=02:44:09 +2026-08-26 14:21:59 | INFO | train rank=0 | step=92650/100000 progress=92.7% loss=0.028011 lr=1.469698e-07 elapsed=34:26:29 avg_step=1.34s recent_step=1.15s eta=02:43:56 +2026-08-26 14:22:11 | INFO | train rank=0 | step=92660/100000 progress=92.7% loss=0.036484 lr=1.465721e-07 elapsed=34:26:41 avg_step=1.34s recent_step=1.22s eta=02:43:42 +2026-08-26 14:22:23 | INFO | train rank=0 | step=92670/100000 progress=92.7% loss=0.049011 lr=1.461750e-07 elapsed=34:26:53 avg_step=1.34s recent_step=1.23s eta=02:43:29 +2026-08-26 14:22:40 | INFO | train rank=0 | step=92680/100000 progress=92.7% loss=0.077057 lr=1.457783e-07 elapsed=34:27:10 avg_step=1.34s recent_step=1.64s eta=02:43:16 +2026-08-26 14:22:52 | INFO | train rank=0 | step=92690/100000 progress=92.7% loss=0.031227 lr=1.453822e-07 elapsed=34:27:22 avg_step=1.34s recent_step=1.20s eta=02:43:02 +2026-08-26 14:23:07 | INFO | train rank=0 | step=92700/100000 progress=92.7% loss=0.055099 lr=1.449867e-07 elapsed=34:27:37 avg_step=1.34s recent_step=1.57s eta=02:42:49 +2026-08-26 14:23:19 | INFO | train rank=0 | step=92710/100000 progress=92.7% loss=0.047882 lr=1.445917e-07 elapsed=34:27:49 avg_step=1.34s recent_step=1.15s eta=02:42:35 +2026-08-26 14:23:34 | INFO | train rank=0 | step=92720/100000 progress=92.7% loss=0.032097 lr=1.441972e-07 elapsed=34:28:03 avg_step=1.34s recent_step=1.46s eta=02:42:22 +2026-08-26 14:23:47 | INFO | train rank=0 | step=92730/100000 progress=92.7% loss=0.052536 lr=1.438032e-07 elapsed=34:28:17 avg_step=1.34s recent_step=1.38s eta=02:42:09 +2026-08-26 14:24:02 | INFO | train rank=0 | step=92740/100000 progress=92.7% loss=0.039810 lr=1.434098e-07 elapsed=34:28:32 avg_step=1.34s recent_step=1.50s eta=02:41:55 +2026-08-26 14:24:14 | INFO | train rank=0 | step=92750/100000 progress=92.8% loss=0.052032 lr=1.430169e-07 elapsed=34:28:44 avg_step=1.34s recent_step=1.20s eta=02:41:42 +2026-08-26 14:24:28 | INFO | train rank=0 | step=92760/100000 progress=92.8% loss=0.029999 lr=1.426245e-07 elapsed=34:28:58 avg_step=1.34s recent_step=1.41s eta=02:41:29 +2026-08-26 14:24:44 | INFO | train rank=0 | step=92770/100000 progress=92.8% loss=0.040436 lr=1.422326e-07 elapsed=34:29:14 avg_step=1.34s recent_step=1.52s eta=02:41:15 +2026-08-26 14:24:56 | INFO | train rank=0 | step=92780/100000 progress=92.8% loss=0.044037 lr=1.418413e-07 elapsed=34:29:26 avg_step=1.34s recent_step=1.21s eta=02:41:02 +2026-08-26 14:25:11 | INFO | train rank=0 | step=92790/100000 progress=92.8% loss=0.091431 lr=1.414506e-07 elapsed=34:29:40 avg_step=1.34s recent_step=1.47s eta=02:40:49 +2026-08-26 14:25:29 | INFO | train rank=0 | step=92800/100000 progress=92.8% loss=0.050476 lr=1.410603e-07 elapsed=34:29:59 avg_step=1.34s recent_step=1.85s eta=02:40:36 +2026-08-26 14:25:40 | INFO | train rank=0 | step=92810/100000 progress=92.8% loss=0.039429 lr=1.406706e-07 elapsed=34:30:10 avg_step=1.34s recent_step=1.14s eta=02:40:22 +2026-08-26 14:25:52 | INFO | train rank=0 | step=92820/100000 progress=92.8% loss=0.055252 lr=1.402814e-07 elapsed=34:30:22 avg_step=1.34s recent_step=1.18s eta=02:40:09 +2026-08-26 14:26:07 | INFO | train rank=0 | step=92830/100000 progress=92.8% loss=0.028221 lr=1.398928e-07 elapsed=34:30:37 avg_step=1.34s recent_step=1.45s eta=02:39:55 +2026-08-26 14:26:19 | INFO | train rank=0 | step=92840/100000 progress=92.8% loss=0.060852 lr=1.395046e-07 elapsed=34:30:49 avg_step=1.34s recent_step=1.20s eta=02:39:42 +2026-08-26 14:26:35 | INFO | train rank=0 | step=92850/100000 progress=92.8% loss=0.042297 lr=1.391170e-07 elapsed=34:31:05 avg_step=1.34s recent_step=1.67s eta=02:39:29 +2026-08-26 14:26:47 | INFO | train rank=0 | step=92860/100000 progress=92.9% loss=0.042152 lr=1.387300e-07 elapsed=34:31:17 avg_step=1.34s recent_step=1.15s eta=02:39:15 +2026-08-26 14:27:01 | INFO | train rank=0 | step=92870/100000 progress=92.9% loss=0.052979 lr=1.383435e-07 elapsed=34:31:31 avg_step=1.34s recent_step=1.41s eta=02:39:02 +2026-08-26 14:27:14 | INFO | train rank=0 | step=92880/100000 progress=92.9% loss=0.046616 lr=1.379575e-07 elapsed=34:31:44 avg_step=1.34s recent_step=1.33s eta=02:38:48 +2026-08-26 14:27:26 | INFO | train rank=0 | step=92890/100000 progress=92.9% loss=0.125122 lr=1.375720e-07 elapsed=34:31:56 avg_step=1.34s recent_step=1.16s eta=02:38:35 +2026-08-26 14:27:39 | INFO | train rank=0 | step=92900/100000 progress=92.9% loss=0.049469 lr=1.371871e-07 elapsed=34:32:09 avg_step=1.34s recent_step=1.34s eta=02:38:22 +2026-08-26 14:27:57 | INFO | train rank=0 | step=92910/100000 progress=92.9% loss=0.044411 lr=1.368027e-07 elapsed=34:32:27 avg_step=1.34s recent_step=1.80s eta=02:38:09 +2026-08-26 14:28:09 | INFO | train rank=0 | step=92920/100000 progress=92.9% loss=0.057846 lr=1.364188e-07 elapsed=34:32:39 avg_step=1.34s recent_step=1.14s eta=02:37:55 +2026-08-26 14:28:20 | INFO | train rank=0 | step=92930/100000 progress=92.9% loss=0.038864 lr=1.360355e-07 elapsed=34:32:50 avg_step=1.34s recent_step=1.16s eta=02:37:41 +2026-08-26 14:28:34 | INFO | train rank=0 | step=92940/100000 progress=92.9% loss=0.046280 lr=1.356527e-07 elapsed=34:33:04 avg_step=1.34s recent_step=1.38s eta=02:37:28 +2026-08-26 14:28:48 | INFO | train rank=0 | step=92950/100000 progress=93.0% loss=0.050385 lr=1.352704e-07 elapsed=34:33:17 avg_step=1.34s recent_step=1.35s eta=02:37:15 +2026-08-26 14:29:01 | INFO | train rank=0 | step=92960/100000 progress=93.0% loss=0.022060 lr=1.348887e-07 elapsed=34:33:30 avg_step=1.34s recent_step=1.30s eta=02:37:01 +2026-08-26 14:29:12 | INFO | train rank=0 | step=92970/100000 progress=93.0% loss=0.034042 lr=1.345074e-07 elapsed=34:33:42 avg_step=1.34s recent_step=1.11s eta=02:36:48 +2026-08-26 14:29:29 | INFO | train rank=0 | step=92980/100000 progress=93.0% loss=0.063812 lr=1.341268e-07 elapsed=34:33:59 avg_step=1.34s recent_step=1.74s eta=02:36:35 +2026-08-26 14:29:41 | INFO | train rank=0 | step=92990/100000 progress=93.0% loss=0.057129 lr=1.337466e-07 elapsed=34:34:11 avg_step=1.34s recent_step=1.18s eta=02:36:21 +2026-08-26 14:29:52 | INFO | train rank=0 | step=93000/100000 progress=93.0% loss=0.076263 lr=1.333670e-07 elapsed=34:34:22 avg_step=1.34s recent_step=1.16s eta=02:36:08 +2026-08-26 14:30:05 | INFO | train rank=0 | step=93010/100000 progress=93.0% loss=0.049255 lr=1.329879e-07 elapsed=34:34:35 avg_step=1.34s recent_step=1.30s eta=02:35:54 +2026-08-26 14:30:17 | INFO | train rank=0 | step=93020/100000 progress=93.0% loss=0.031807 lr=1.326094e-07 elapsed=34:34:47 avg_step=1.34s recent_step=1.19s eta=02:35:41 +2026-08-26 14:30:36 | INFO | train rank=0 | step=93030/100000 progress=93.0% loss=0.043289 lr=1.322314e-07 elapsed=34:35:06 avg_step=1.34s recent_step=1.87s eta=02:35:28 +2026-08-26 14:30:48 | INFO | train rank=0 | step=93040/100000 progress=93.0% loss=0.037384 lr=1.318539e-07 elapsed=34:35:18 avg_step=1.34s recent_step=1.21s eta=02:35:14 +2026-08-26 14:31:03 | INFO | train rank=0 | step=93050/100000 progress=93.0% loss=0.030708 lr=1.314769e-07 elapsed=34:35:32 avg_step=1.34s recent_step=1.46s eta=02:35:01 +2026-08-26 14:31:14 | INFO | train rank=0 | step=93060/100000 progress=93.1% loss=0.063232 lr=1.311005e-07 elapsed=34:35:44 avg_step=1.34s recent_step=1.19s eta=02:34:47 +2026-08-26 14:31:32 | INFO | train rank=0 | step=93070/100000 progress=93.1% loss=0.115356 lr=1.307246e-07 elapsed=34:36:02 avg_step=1.34s recent_step=1.76s eta=02:34:34 +2026-08-26 14:31:45 | INFO | train rank=0 | step=93080/100000 progress=93.1% loss=0.078674 lr=1.303493e-07 elapsed=34:36:15 avg_step=1.34s recent_step=1.27s eta=02:34:21 +2026-08-26 14:31:56 | INFO | train rank=0 | step=93090/100000 progress=93.1% loss=0.043030 lr=1.299745e-07 elapsed=34:36:26 avg_step=1.34s recent_step=1.15s eta=02:34:07 +2026-08-26 14:32:08 | INFO | train rank=0 | step=93100/100000 progress=93.1% loss=0.035370 lr=1.296002e-07 elapsed=34:36:38 avg_step=1.34s recent_step=1.14s eta=02:33:54 +2026-08-26 14:32:19 | INFO | train rank=0 | step=93110/100000 progress=93.1% loss=0.035965 lr=1.292264e-07 elapsed=34:36:49 avg_step=1.34s recent_step=1.14s eta=02:33:40 +2026-08-26 14:32:31 | INFO | train rank=0 | step=93120/100000 progress=93.1% loss=0.044495 lr=1.288532e-07 elapsed=34:37:01 avg_step=1.34s recent_step=1.19s eta=02:33:27 +2026-08-26 14:32:45 | INFO | train rank=0 | step=93130/100000 progress=93.1% loss=0.047638 lr=1.284805e-07 elapsed=34:37:15 avg_step=1.34s recent_step=1.44s eta=02:33:14 +2026-08-26 14:33:01 | INFO | train rank=0 | step=93140/100000 progress=93.1% loss=0.034637 lr=1.281084e-07 elapsed=34:37:30 avg_step=1.34s recent_step=1.52s eta=02:33:00 +2026-08-26 14:33:15 | INFO | train rank=0 | step=93150/100000 progress=93.2% loss=0.043045 lr=1.277367e-07 elapsed=34:37:45 avg_step=1.34s recent_step=1.48s eta=02:32:47 +2026-08-26 14:33:27 | INFO | train rank=0 | step=93160/100000 progress=93.2% loss=0.040184 lr=1.273656e-07 elapsed=34:37:56 avg_step=1.34s recent_step=1.12s eta=02:32:34 +2026-08-26 14:33:44 | INFO | train rank=0 | step=93170/100000 progress=93.2% loss=0.042206 lr=1.269951e-07 elapsed=34:38:14 avg_step=1.34s recent_step=1.76s eta=02:32:20 +2026-08-26 14:33:58 | INFO | train rank=0 | step=93180/100000 progress=93.2% loss=0.053345 lr=1.266251e-07 elapsed=34:38:28 avg_step=1.34s recent_step=1.41s eta=02:32:07 +2026-08-26 14:34:12 | INFO | train rank=0 | step=93190/100000 progress=93.2% loss=0.035995 lr=1.262556e-07 elapsed=34:38:42 avg_step=1.34s recent_step=1.39s eta=02:31:54 +2026-08-26 14:34:27 | INFO | train rank=0 | step=93200/100000 progress=93.2% loss=0.056244 lr=1.258866e-07 elapsed=34:38:57 avg_step=1.34s recent_step=1.53s eta=02:31:41 +2026-08-26 14:34:39 | INFO | train rank=0 | step=93210/100000 progress=93.2% loss=0.041916 lr=1.255182e-07 elapsed=34:39:09 avg_step=1.34s recent_step=1.15s eta=02:31:27 +2026-08-26 14:34:53 | INFO | train rank=0 | step=93220/100000 progress=93.2% loss=0.025879 lr=1.251503e-07 elapsed=34:39:23 avg_step=1.34s recent_step=1.44s eta=02:31:14 +2026-08-26 14:35:07 | INFO | train rank=0 | step=93230/100000 progress=93.2% loss=0.042755 lr=1.247829e-07 elapsed=34:39:37 avg_step=1.34s recent_step=1.38s eta=02:31:00 +2026-08-26 14:35:19 | INFO | train rank=0 | step=93240/100000 progress=93.2% loss=0.045525 lr=1.244161e-07 elapsed=34:39:49 avg_step=1.34s recent_step=1.19s eta=02:30:47 +2026-08-26 14:35:33 | INFO | train rank=0 | step=93250/100000 progress=93.2% loss=0.050308 lr=1.240498e-07 elapsed=34:40:03 avg_step=1.34s recent_step=1.39s eta=02:30:34 +2026-08-26 14:35:44 | INFO | train rank=0 | step=93260/100000 progress=93.3% loss=0.078094 lr=1.236840e-07 elapsed=34:40:14 avg_step=1.34s recent_step=1.13s eta=02:30:20 +2026-08-26 14:36:00 | INFO | train rank=0 | step=93270/100000 progress=93.3% loss=0.040672 lr=1.233188e-07 elapsed=34:40:30 avg_step=1.34s recent_step=1.59s eta=02:30:07 +2026-08-26 14:36:12 | INFO | train rank=0 | step=93280/100000 progress=93.3% loss=0.068451 lr=1.229541e-07 elapsed=34:40:42 avg_step=1.34s recent_step=1.19s eta=02:29:53 +2026-08-26 14:36:25 | INFO | train rank=0 | step=93290/100000 progress=93.3% loss=0.038925 lr=1.225900e-07 elapsed=34:40:55 avg_step=1.34s recent_step=1.28s eta=02:29:40 +2026-08-26 14:36:36 | INFO | train rank=0 | step=93300/100000 progress=93.3% loss=0.033020 lr=1.222263e-07 elapsed=34:41:06 avg_step=1.34s recent_step=1.15s eta=02:29:26 +2026-08-26 14:36:48 | INFO | train rank=0 | step=93310/100000 progress=93.3% loss=0.044052 lr=1.218632e-07 elapsed=34:41:17 avg_step=1.34s recent_step=1.12s eta=02:29:13 +2026-08-26 14:36:59 | INFO | train rank=0 | step=93320/100000 progress=93.3% loss=0.046776 lr=1.215007e-07 elapsed=34:41:29 avg_step=1.34s recent_step=1.14s eta=02:28:59 +2026-08-26 14:37:14 | INFO | train rank=0 | step=93330/100000 progress=93.3% loss=0.041901 lr=1.211386e-07 elapsed=34:41:44 avg_step=1.34s recent_step=1.50s eta=02:28:46 +2026-08-26 14:37:25 | INFO | train rank=0 | step=93340/100000 progress=93.3% loss=0.031326 lr=1.207772e-07 elapsed=34:41:55 avg_step=1.34s recent_step=1.15s eta=02:28:33 +2026-08-26 14:37:41 | INFO | train rank=0 | step=93350/100000 progress=93.3% loss=0.062836 lr=1.204162e-07 elapsed=34:42:11 avg_step=1.34s recent_step=1.55s eta=02:28:19 +2026-08-26 14:37:56 | INFO | train rank=0 | step=93360/100000 progress=93.4% loss=0.080383 lr=1.200558e-07 elapsed=34:42:25 avg_step=1.34s recent_step=1.47s eta=02:28:06 +2026-08-26 14:38:07 | INFO | train rank=0 | step=93370/100000 progress=93.4% loss=0.065033 lr=1.196959e-07 elapsed=34:42:37 avg_step=1.34s recent_step=1.13s eta=02:27:52 +2026-08-26 14:38:21 | INFO | train rank=0 | step=93380/100000 progress=93.4% loss=0.062836 lr=1.193365e-07 elapsed=34:42:51 avg_step=1.34s recent_step=1.38s eta=02:27:39 +2026-08-26 14:38:39 | INFO | train rank=0 | step=93390/100000 progress=93.4% loss=0.054626 lr=1.189777e-07 elapsed=34:43:09 avg_step=1.34s recent_step=1.80s eta=02:27:26 +2026-08-26 14:38:50 | INFO | train rank=0 | step=93400/100000 progress=93.4% loss=0.065735 lr=1.186194e-07 elapsed=34:43:20 avg_step=1.34s recent_step=1.17s eta=02:27:13 +2026-08-26 14:39:05 | INFO | train rank=0 | step=93410/100000 progress=93.4% loss=0.063553 lr=1.182616e-07 elapsed=34:43:35 avg_step=1.34s recent_step=1.48s eta=02:26:59 +2026-08-26 14:39:16 | INFO | train rank=0 | step=93420/100000 progress=93.4% loss=0.050652 lr=1.179044e-07 elapsed=34:43:46 avg_step=1.34s recent_step=1.12s eta=02:26:46 +2026-08-26 14:39:28 | INFO | train rank=0 | step=93430/100000 progress=93.4% loss=0.038673 lr=1.175477e-07 elapsed=34:43:58 avg_step=1.34s recent_step=1.18s eta=02:26:32 +2026-08-26 14:39:39 | INFO | train rank=0 | step=93440/100000 progress=93.4% loss=0.063751 lr=1.171916e-07 elapsed=34:44:09 avg_step=1.34s recent_step=1.09s eta=02:26:19 +2026-08-26 14:39:56 | INFO | train rank=0 | step=93450/100000 progress=93.5% loss=0.043633 lr=1.168359e-07 elapsed=34:44:26 avg_step=1.34s recent_step=1.69s eta=02:26:06 +2026-08-26 14:40:10 | INFO | train rank=0 | step=93460/100000 progress=93.5% loss=0.098877 lr=1.164809e-07 elapsed=34:44:40 avg_step=1.34s recent_step=1.38s eta=02:25:52 +2026-08-26 14:40:22 | INFO | train rank=0 | step=93470/100000 progress=93.5% loss=0.049313 lr=1.161263e-07 elapsed=34:44:51 avg_step=1.34s recent_step=1.17s eta=02:25:39 +2026-08-26 14:40:33 | INFO | train rank=0 | step=93480/100000 progress=93.5% loss=0.050751 lr=1.157723e-07 elapsed=34:45:03 avg_step=1.34s recent_step=1.19s eta=02:25:25 +2026-08-26 14:40:47 | INFO | train rank=0 | step=93490/100000 progress=93.5% loss=0.068390 lr=1.154188e-07 elapsed=34:45:17 avg_step=1.34s recent_step=1.33s eta=02:25:12 +2026-08-26 14:41:03 | INFO | train rank=0 | step=93500/100000 progress=93.5% loss=0.055420 lr=1.150658e-07 elapsed=34:45:33 avg_step=1.34s recent_step=1.59s eta=02:24:59 +2026-08-26 14:41:14 | INFO | train rank=0 | step=93510/100000 progress=93.5% loss=0.043732 lr=1.147134e-07 elapsed=34:45:44 avg_step=1.34s recent_step=1.16s eta=02:24:45 +2026-08-26 14:41:28 | INFO | train rank=0 | step=93520/100000 progress=93.5% loss=0.036400 lr=1.143615e-07 elapsed=34:45:57 avg_step=1.34s recent_step=1.33s eta=02:24:32 +2026-08-26 14:41:43 | INFO | train rank=0 | step=93530/100000 progress=93.5% loss=0.049942 lr=1.140102e-07 elapsed=34:46:13 avg_step=1.34s recent_step=1.53s eta=02:24:18 +2026-08-26 14:41:55 | INFO | train rank=0 | step=93540/100000 progress=93.5% loss=0.040794 lr=1.136594e-07 elapsed=34:46:25 avg_step=1.34s recent_step=1.20s eta=02:24:05 +2026-08-26 14:42:09 | INFO | train rank=0 | step=93550/100000 progress=93.5% loss=0.033829 lr=1.133091e-07 elapsed=34:46:39 avg_step=1.34s recent_step=1.44s eta=02:23:52 +2026-08-26 14:42:25 | INFO | train rank=0 | step=93560/100000 progress=93.6% loss=0.046417 lr=1.129594e-07 elapsed=34:46:55 avg_step=1.34s recent_step=1.55s eta=02:23:38 +2026-08-26 14:42:36 | INFO | train rank=0 | step=93570/100000 progress=93.6% loss=0.026344 lr=1.126101e-07 elapsed=34:47:06 avg_step=1.34s recent_step=1.13s eta=02:23:25 +2026-08-26 14:42:48 | INFO | train rank=0 | step=93580/100000 progress=93.6% loss=0.039520 lr=1.122615e-07 elapsed=34:47:18 avg_step=1.34s recent_step=1.18s eta=02:23:11 +2026-08-26 14:42:59 | INFO | train rank=0 | step=93590/100000 progress=93.6% loss=0.040146 lr=1.119133e-07 elapsed=34:47:29 avg_step=1.34s recent_step=1.14s eta=02:22:58 +2026-08-26 14:43:16 | INFO | train rank=0 | step=93600/100000 progress=93.6% loss=0.115356 lr=1.115657e-07 elapsed=34:47:45 avg_step=1.34s recent_step=1.61s eta=02:22:45 +2026-08-26 14:43:27 | INFO | train rank=0 | step=93610/100000 progress=93.6% loss=0.047745 lr=1.112186e-07 elapsed=34:47:57 avg_step=1.34s recent_step=1.19s eta=02:22:31 +2026-08-26 14:43:41 | INFO | train rank=0 | step=93620/100000 progress=93.6% loss=0.033882 lr=1.108721e-07 elapsed=34:48:11 avg_step=1.34s recent_step=1.39s eta=02:22:18 +2026-08-26 14:43:55 | INFO | train rank=0 | step=93630/100000 progress=93.6% loss=0.032772 lr=1.105261e-07 elapsed=34:48:25 avg_step=1.34s recent_step=1.33s eta=02:22:04 +2026-08-26 14:44:09 | INFO | train rank=0 | step=93640/100000 progress=93.6% loss=0.076202 lr=1.101806e-07 elapsed=34:48:39 avg_step=1.34s recent_step=1.44s eta=02:21:51 +2026-08-26 14:44:22 | INFO | train rank=0 | step=93650/100000 progress=93.7% loss=0.040390 lr=1.098357e-07 elapsed=34:48:52 avg_step=1.34s recent_step=1.33s eta=02:21:38 +2026-08-26 14:44:34 | INFO | train rank=0 | step=93660/100000 progress=93.7% loss=0.047905 lr=1.094913e-07 elapsed=34:49:04 avg_step=1.34s recent_step=1.19s eta=02:21:24 +2026-08-26 14:44:49 | INFO | train rank=0 | step=93670/100000 progress=93.7% loss=0.064423 lr=1.091474e-07 elapsed=34:49:19 avg_step=1.34s recent_step=1.51s eta=02:21:11 +2026-08-26 14:45:01 | INFO | train rank=0 | step=93680/100000 progress=93.7% loss=0.043747 lr=1.088041e-07 elapsed=34:49:30 avg_step=1.34s recent_step=1.13s eta=02:20:57 +2026-08-26 14:45:16 | INFO | train rank=0 | step=93690/100000 progress=93.7% loss=0.023033 lr=1.084613e-07 elapsed=34:49:46 avg_step=1.34s recent_step=1.55s eta=02:20:44 +2026-08-26 14:45:29 | INFO | train rank=0 | step=93700/100000 progress=93.7% loss=0.034866 lr=1.081190e-07 elapsed=34:49:59 avg_step=1.34s recent_step=1.28s eta=02:20:31 +2026-08-26 14:45:46 | INFO | train rank=0 | step=93710/100000 progress=93.7% loss=0.102234 lr=1.077773e-07 elapsed=34:50:15 avg_step=1.34s recent_step=1.68s eta=02:20:18 +2026-08-26 14:45:57 | INFO | train rank=0 | step=93720/100000 progress=93.7% loss=0.049728 lr=1.074361e-07 elapsed=34:50:27 avg_step=1.34s recent_step=1.19s eta=02:20:04 +2026-08-26 14:46:15 | INFO | train rank=0 | step=93730/100000 progress=93.7% loss=0.040009 lr=1.070955e-07 elapsed=34:50:44 avg_step=1.34s recent_step=1.71s eta=02:19:51 +2026-08-26 14:46:26 | INFO | train rank=0 | step=93740/100000 progress=93.7% loss=0.046356 lr=1.067554e-07 elapsed=34:50:56 avg_step=1.34s recent_step=1.18s eta=02:19:38 +2026-08-26 14:46:38 | INFO | train rank=0 | step=93750/100000 progress=93.8% loss=0.030289 lr=1.064158e-07 elapsed=34:51:08 avg_step=1.34s recent_step=1.14s eta=02:19:24 +2026-08-26 14:46:51 | INFO | train rank=0 | step=93760/100000 progress=93.8% loss=0.034515 lr=1.060767e-07 elapsed=34:51:20 avg_step=1.34s recent_step=1.28s eta=02:19:11 +2026-08-26 14:47:06 | INFO | train rank=0 | step=93770/100000 progress=93.8% loss=0.017311 lr=1.057382e-07 elapsed=34:51:36 avg_step=1.34s recent_step=1.59s eta=02:18:57 +2026-08-26 14:47:18 | INFO | train rank=0 | step=93780/100000 progress=93.8% loss=0.064819 lr=1.054002e-07 elapsed=34:51:48 avg_step=1.34s recent_step=1.14s eta=02:18:44 +2026-08-26 14:47:30 | INFO | train rank=0 | step=93790/100000 progress=93.8% loss=0.056564 lr=1.050628e-07 elapsed=34:52:00 avg_step=1.34s recent_step=1.24s eta=02:18:30 +2026-08-26 14:47:45 | INFO | train rank=0 | step=93800/100000 progress=93.8% loss=0.039963 lr=1.047259e-07 elapsed=34:52:15 avg_step=1.34s recent_step=1.45s eta=02:18:17 +2026-08-26 14:47:59 | INFO | train rank=0 | step=93810/100000 progress=93.8% loss=0.061211 lr=1.043895e-07 elapsed=34:52:29 avg_step=1.34s recent_step=1.39s eta=02:18:04 +2026-08-26 14:48:11 | INFO | train rank=0 | step=93820/100000 progress=93.8% loss=0.073761 lr=1.040537e-07 elapsed=34:52:41 avg_step=1.34s recent_step=1.24s eta=02:17:50 +2026-08-26 14:48:27 | INFO | train rank=0 | step=93830/100000 progress=93.8% loss=0.036720 lr=1.037184e-07 elapsed=34:52:57 avg_step=1.34s recent_step=1.59s eta=02:17:37 +2026-08-26 14:48:42 | INFO | train rank=0 | step=93840/100000 progress=93.8% loss=0.037140 lr=1.033836e-07 elapsed=34:53:12 avg_step=1.34s recent_step=1.51s eta=02:17:24 +2026-08-26 14:48:54 | INFO | train rank=0 | step=93850/100000 progress=93.8% loss=0.039017 lr=1.030494e-07 elapsed=34:53:24 avg_step=1.34s recent_step=1.20s eta=02:17:10 +2026-08-26 14:49:06 | INFO | train rank=0 | step=93860/100000 progress=93.9% loss=0.046669 lr=1.027157e-07 elapsed=34:53:36 avg_step=1.34s recent_step=1.20s eta=02:16:57 +2026-08-26 14:49:19 | INFO | train rank=0 | step=93870/100000 progress=93.9% loss=0.045898 lr=1.023825e-07 elapsed=34:53:49 avg_step=1.34s recent_step=1.32s eta=02:16:43 +2026-08-26 14:49:31 | INFO | train rank=0 | step=93880/100000 progress=93.9% loss=0.051743 lr=1.020499e-07 elapsed=34:54:00 avg_step=1.34s recent_step=1.15s eta=02:16:30 +2026-08-26 14:49:46 | INFO | train rank=0 | step=93890/100000 progress=93.9% loss=0.061661 lr=1.017178e-07 elapsed=34:54:16 avg_step=1.34s recent_step=1.55s eta=02:16:17 +2026-08-26 14:50:01 | INFO | train rank=0 | step=93900/100000 progress=93.9% loss=0.044006 lr=1.013862e-07 elapsed=34:54:31 avg_step=1.34s recent_step=1.53s eta=02:16:03 +2026-08-26 14:50:13 | INFO | train rank=0 | step=93910/100000 progress=93.9% loss=0.040665 lr=1.010552e-07 elapsed=34:54:43 avg_step=1.34s recent_step=1.15s eta=02:15:50 +2026-08-26 14:50:26 | INFO | train rank=0 | step=93920/100000 progress=93.9% loss=0.069031 lr=1.007247e-07 elapsed=34:54:56 avg_step=1.34s recent_step=1.36s eta=02:15:37 +2026-08-26 14:50:39 | INFO | train rank=0 | step=93930/100000 progress=93.9% loss=0.066315 lr=1.003948e-07 elapsed=34:55:09 avg_step=1.34s recent_step=1.24s eta=02:15:23 +2026-08-26 14:50:53 | INFO | train rank=0 | step=93940/100000 progress=93.9% loss=0.040009 lr=1.000654e-07 elapsed=34:55:22 avg_step=1.34s recent_step=1.38s eta=02:15:10 +2026-08-26 14:51:04 | INFO | train rank=0 | step=93950/100000 progress=94.0% loss=0.042801 lr=9.973651e-08 elapsed=34:55:34 avg_step=1.34s recent_step=1.17s eta=02:14:56 +2026-08-26 14:51:16 | INFO | train rank=0 | step=93960/100000 progress=94.0% loss=0.046371 lr=9.940817e-08 elapsed=34:55:46 avg_step=1.34s recent_step=1.17s eta=02:14:43 +2026-08-26 14:51:28 | INFO | train rank=0 | step=93970/100000 progress=94.0% loss=0.074615 lr=9.908036e-08 elapsed=34:55:58 avg_step=1.34s recent_step=1.24s eta=02:14:29 +2026-08-26 14:51:43 | INFO | train rank=0 | step=93980/100000 progress=94.0% loss=0.045914 lr=9.875310e-08 elapsed=34:56:13 avg_step=1.34s recent_step=1.50s eta=02:14:16 +2026-08-26 14:51:58 | INFO | train rank=0 | step=93990/100000 progress=94.0% loss=0.043121 lr=9.842637e-08 elapsed=34:56:28 avg_step=1.34s recent_step=1.44s eta=02:14:03 +2026-08-26 14:52:11 | INFO | train rank=0 | step=94000/100000 progress=94.0% loss=0.041962 lr=9.810017e-08 elapsed=34:56:41 avg_step=1.34s recent_step=1.29s eta=02:13:49 +2026-08-26 14:52:28 | INFO | train rank=0 | step=94010/100000 progress=94.0% loss=0.042770 lr=9.777451e-08 elapsed=34:56:58 avg_step=1.34s recent_step=1.71s eta=02:13:36 +2026-08-26 14:52:39 | INFO | train rank=0 | step=94020/100000 progress=94.0% loss=0.037491 lr=9.744939e-08 elapsed=34:57:09 avg_step=1.34s recent_step=1.16s eta=02:13:23 +2026-08-26 14:52:51 | INFO | train rank=0 | step=94030/100000 progress=94.0% loss=0.032440 lr=9.712480e-08 elapsed=34:57:21 avg_step=1.34s recent_step=1.18s eta=02:13:09 +2026-08-26 14:53:03 | INFO | train rank=0 | step=94040/100000 progress=94.0% loss=0.046814 lr=9.680075e-08 elapsed=34:57:33 avg_step=1.34s recent_step=1.19s eta=02:12:56 +2026-08-26 14:53:15 | INFO | train rank=0 | step=94050/100000 progress=94.0% loss=0.028053 lr=9.647724e-08 elapsed=34:57:45 avg_step=1.34s recent_step=1.16s eta=02:12:42 +2026-08-26 14:53:30 | INFO | train rank=0 | step=94060/100000 progress=94.1% loss=0.049164 lr=9.615426e-08 elapsed=34:58:00 avg_step=1.34s recent_step=1.57s eta=02:12:29 +2026-08-26 14:53:45 | INFO | train rank=0 | step=94070/100000 progress=94.1% loss=0.037643 lr=9.583182e-08 elapsed=34:58:15 avg_step=1.34s recent_step=1.43s eta=02:12:16 +2026-08-26 14:53:56 | INFO | train rank=0 | step=94080/100000 progress=94.1% loss=0.055313 lr=9.550991e-08 elapsed=34:58:26 avg_step=1.34s recent_step=1.17s eta=02:12:02 +2026-08-26 14:54:08 | INFO | train rank=0 | step=94090/100000 progress=94.1% loss=0.051941 lr=9.518854e-08 elapsed=34:58:38 avg_step=1.34s recent_step=1.20s eta=02:11:49 +2026-08-26 14:54:22 | INFO | train rank=0 | step=94100/100000 progress=94.1% loss=0.055511 lr=9.486771e-08 elapsed=34:58:52 avg_step=1.34s recent_step=1.38s eta=02:11:35 +2026-08-26 14:54:34 | INFO | train rank=0 | step=94110/100000 progress=94.1% loss=0.103638 lr=9.454741e-08 elapsed=34:59:04 avg_step=1.34s recent_step=1.15s eta=02:11:22 +2026-08-26 14:54:48 | INFO | train rank=0 | step=94120/100000 progress=94.1% loss=0.069122 lr=9.422765e-08 elapsed=34:59:18 avg_step=1.34s recent_step=1.42s eta=02:11:09 +2026-08-26 14:55:02 | INFO | train rank=0 | step=94130/100000 progress=94.1% loss=0.038605 lr=9.390843e-08 elapsed=34:59:32 avg_step=1.34s recent_step=1.39s eta=02:10:55 +2026-08-26 14:55:16 | INFO | train rank=0 | step=94140/100000 progress=94.1% loss=0.075409 lr=9.358974e-08 elapsed=34:59:46 avg_step=1.34s recent_step=1.41s eta=02:10:42 +2026-08-26 14:55:31 | INFO | train rank=0 | step=94150/100000 progress=94.2% loss=0.029175 lr=9.327159e-08 elapsed=35:00:01 avg_step=1.34s recent_step=1.55s eta=02:10:29 +2026-08-26 14:55:43 | INFO | train rank=0 | step=94160/100000 progress=94.2% loss=0.034737 lr=9.295398e-08 elapsed=35:00:13 avg_step=1.34s recent_step=1.17s eta=02:10:15 +2026-08-26 14:55:56 | INFO | train rank=0 | step=94170/100000 progress=94.2% loss=0.037323 lr=9.263690e-08 elapsed=35:00:26 avg_step=1.34s recent_step=1.32s eta=02:10:02 +2026-08-26 14:56:11 | INFO | train rank=0 | step=94180/100000 progress=94.2% loss=0.041237 lr=9.232036e-08 elapsed=35:00:41 avg_step=1.34s recent_step=1.51s eta=02:09:48 +2026-08-26 14:56:29 | INFO | train rank=0 | step=94190/100000 progress=94.2% loss=0.061035 lr=9.200436e-08 elapsed=35:00:59 avg_step=1.34s recent_step=1.80s eta=02:09:35 +2026-08-26 14:56:41 | INFO | train rank=0 | step=94200/100000 progress=94.2% loss=0.041183 lr=9.168889e-08 elapsed=35:01:11 avg_step=1.34s recent_step=1.14s eta=02:09:22 +2026-08-26 14:56:55 | INFO | train rank=0 | step=94210/100000 progress=94.2% loss=0.038887 lr=9.137396e-08 elapsed=35:01:24 avg_step=1.34s recent_step=1.37s eta=02:09:08 +2026-08-26 14:57:09 | INFO | train rank=0 | step=94220/100000 progress=94.2% loss=0.036011 lr=9.105957e-08 elapsed=35:01:39 avg_step=1.34s recent_step=1.47s eta=02:08:55 +2026-08-26 14:57:28 | INFO | train rank=0 | step=94230/100000 progress=94.2% loss=0.047302 lr=9.074571e-08 elapsed=35:01:58 avg_step=1.34s recent_step=1.90s eta=02:08:42 +2026-08-26 14:57:43 | INFO | train rank=0 | step=94240/100000 progress=94.2% loss=0.047623 lr=9.043239e-08 elapsed=35:02:13 avg_step=1.34s recent_step=1.47s eta=02:08:29 +2026-08-26 14:57:57 | INFO | train rank=0 | step=94250/100000 progress=94.2% loss=0.036560 lr=9.011961e-08 elapsed=35:02:27 avg_step=1.34s recent_step=1.39s eta=02:08:15 +2026-08-26 14:58:08 | INFO | train rank=0 | step=94260/100000 progress=94.3% loss=0.038063 lr=8.980736e-08 elapsed=35:02:38 avg_step=1.34s recent_step=1.18s eta=02:08:02 +2026-08-26 14:58:22 | INFO | train rank=0 | step=94270/100000 progress=94.3% loss=0.047417 lr=8.949565e-08 elapsed=35:02:52 avg_step=1.34s recent_step=1.37s eta=02:07:49 +2026-08-26 14:58:34 | INFO | train rank=0 | step=94280/100000 progress=94.3% loss=0.079926 lr=8.918448e-08 elapsed=35:03:04 avg_step=1.34s recent_step=1.15s eta=02:07:35 +2026-08-26 14:58:46 | INFO | train rank=0 | step=94290/100000 progress=94.3% loss=0.024605 lr=8.887384e-08 elapsed=35:03:16 avg_step=1.34s recent_step=1.19s eta=02:07:22 +2026-08-26 14:58:58 | INFO | train rank=0 | step=94300/100000 progress=94.3% loss=0.032722 lr=8.856375e-08 elapsed=35:03:28 avg_step=1.34s recent_step=1.27s eta=02:07:08 +2026-08-26 14:59:13 | INFO | train rank=0 | step=94310/100000 progress=94.3% loss=0.036224 lr=8.825419e-08 elapsed=35:03:43 avg_step=1.34s recent_step=1.46s eta=02:06:55 +2026-08-26 14:59:25 | INFO | train rank=0 | step=94320/100000 progress=94.3% loss=0.047485 lr=8.794516e-08 elapsed=35:03:55 avg_step=1.34s recent_step=1.24s eta=02:06:41 +2026-08-26 14:59:43 | INFO | train rank=0 | step=94330/100000 progress=94.3% loss=0.039886 lr=8.763668e-08 elapsed=35:04:13 avg_step=1.34s recent_step=1.77s eta=02:06:28 +2026-08-26 14:59:55 | INFO | train rank=0 | step=94340/100000 progress=94.3% loss=0.064087 lr=8.732873e-08 elapsed=35:04:25 avg_step=1.34s recent_step=1.20s eta=02:06:15 +2026-08-26 15:00:07 | INFO | train rank=0 | step=94350/100000 progress=94.3% loss=0.035732 lr=8.702131e-08 elapsed=35:04:36 avg_step=1.34s recent_step=1.15s eta=02:06:01 +2026-08-26 15:00:23 | INFO | train rank=0 | step=94360/100000 progress=94.4% loss=0.050423 lr=8.671444e-08 elapsed=35:04:52 avg_step=1.34s recent_step=1.59s eta=02:05:48 +2026-08-26 15:00:36 | INFO | train rank=0 | step=94370/100000 progress=94.4% loss=0.032013 lr=8.640810e-08 elapsed=35:05:06 avg_step=1.34s recent_step=1.33s eta=02:05:35 +2026-08-26 15:00:51 | INFO | train rank=0 | step=94380/100000 progress=94.4% loss=0.032372 lr=8.610230e-08 elapsed=35:05:21 avg_step=1.34s recent_step=1.52s eta=02:05:21 +2026-08-26 15:01:06 | INFO | train rank=0 | step=94390/100000 progress=94.4% loss=0.060974 lr=8.579704e-08 elapsed=35:05:36 avg_step=1.34s recent_step=1.47s eta=02:05:08 +2026-08-26 15:01:19 | INFO | train rank=0 | step=94400/100000 progress=94.4% loss=0.030655 lr=8.549231e-08 elapsed=35:05:49 avg_step=1.34s recent_step=1.35s eta=02:04:55 +2026-08-26 15:01:35 | INFO | train rank=0 | step=94410/100000 progress=94.4% loss=0.034256 lr=8.518813e-08 elapsed=35:06:05 avg_step=1.34s recent_step=1.63s eta=02:04:42 +2026-08-26 15:01:47 | INFO | train rank=0 | step=94420/100000 progress=94.4% loss=0.061890 lr=8.488448e-08 elapsed=35:06:17 avg_step=1.34s recent_step=1.14s eta=02:04:28 +2026-08-26 15:02:03 | INFO | train rank=0 | step=94430/100000 progress=94.4% loss=0.090393 lr=8.458136e-08 elapsed=35:06:33 avg_step=1.34s recent_step=1.58s eta=02:04:15 +2026-08-26 15:02:14 | INFO | train rank=0 | step=94440/100000 progress=94.4% loss=0.043228 lr=8.427879e-08 elapsed=35:06:44 avg_step=1.34s recent_step=1.10s eta=02:04:01 +2026-08-26 15:02:26 | INFO | train rank=0 | step=94450/100000 progress=94.5% loss=0.048477 lr=8.397675e-08 elapsed=35:06:56 avg_step=1.34s recent_step=1.26s eta=02:03:48 +2026-08-26 15:02:43 | INFO | train rank=0 | step=94460/100000 progress=94.5% loss=0.038162 lr=8.367525e-08 elapsed=35:07:13 avg_step=1.34s recent_step=1.65s eta=02:03:35 +2026-08-26 15:02:54 | INFO | train rank=0 | step=94470/100000 progress=94.5% loss=0.047173 lr=8.337429e-08 elapsed=35:07:24 avg_step=1.34s recent_step=1.15s eta=02:03:21 +2026-08-26 15:03:06 | INFO | train rank=0 | step=94480/100000 progress=94.5% loss=0.024986 lr=8.307386e-08 elapsed=35:07:36 avg_step=1.34s recent_step=1.14s eta=02:03:08 +2026-08-26 15:03:18 | INFO | train rank=0 | step=94490/100000 progress=94.5% loss=0.052704 lr=8.277398e-08 elapsed=35:07:48 avg_step=1.34s recent_step=1.26s eta=02:02:54 +2026-08-26 15:03:31 | INFO | train rank=0 | step=94500/100000 progress=94.5% loss=0.047714 lr=8.247463e-08 elapsed=35:08:01 avg_step=1.34s recent_step=1.27s eta=02:02:41 +2026-08-26 15:03:42 | INFO | train rank=0 | step=94510/100000 progress=94.5% loss=0.093414 lr=8.217581e-08 elapsed=35:08:12 avg_step=1.34s recent_step=1.13s eta=02:02:27 +2026-08-26 15:03:56 | INFO | train rank=0 | step=94520/100000 progress=94.5% loss=0.057861 lr=8.187754e-08 elapsed=35:08:26 avg_step=1.34s recent_step=1.35s eta=02:02:14 +2026-08-26 15:04:11 | INFO | train rank=0 | step=94530/100000 progress=94.5% loss=0.036072 lr=8.157980e-08 elapsed=35:08:41 avg_step=1.34s recent_step=1.48s eta=02:02:01 +2026-08-26 15:04:22 | INFO | train rank=0 | step=94540/100000 progress=94.5% loss=0.029259 lr=8.128261e-08 elapsed=35:08:52 avg_step=1.34s recent_step=1.15s eta=02:01:47 +2026-08-26 15:04:36 | INFO | train rank=0 | step=94550/100000 progress=94.5% loss=0.057709 lr=8.098595e-08 elapsed=35:09:06 avg_step=1.34s recent_step=1.42s eta=02:01:34 +2026-08-26 15:04:51 | INFO | train rank=0 | step=94560/100000 progress=94.6% loss=0.063889 lr=8.068982e-08 elapsed=35:09:21 avg_step=1.34s recent_step=1.44s eta=02:01:21 +2026-08-26 15:05:03 | INFO | train rank=0 | step=94570/100000 progress=94.6% loss=0.054459 lr=8.039424e-08 elapsed=35:09:32 avg_step=1.34s recent_step=1.17s eta=02:01:07 +2026-08-26 15:05:17 | INFO | train rank=0 | step=94580/100000 progress=94.6% loss=0.044930 lr=8.009919e-08 elapsed=35:09:47 avg_step=1.34s recent_step=1.48s eta=02:00:54 +2026-08-26 15:05:31 | INFO | train rank=0 | step=94590/100000 progress=94.6% loss=0.034599 lr=7.980468e-08 elapsed=35:10:01 avg_step=1.34s recent_step=1.37s eta=02:00:40 +2026-08-26 15:05:46 | INFO | train rank=0 | step=94600/100000 progress=94.6% loss=0.045135 lr=7.951071e-08 elapsed=35:10:16 avg_step=1.34s recent_step=1.50s eta=02:00:27 +2026-08-26 15:05:58 | INFO | train rank=0 | step=94610/100000 progress=94.6% loss=0.045364 lr=7.921728e-08 elapsed=35:10:27 avg_step=1.34s recent_step=1.16s eta=02:00:14 +2026-08-26 15:06:13 | INFO | train rank=0 | step=94620/100000 progress=94.6% loss=0.058228 lr=7.892439e-08 elapsed=35:10:42 avg_step=1.34s recent_step=1.50s eta=02:00:00 +2026-08-26 15:06:25 | INFO | train rank=0 | step=94630/100000 progress=94.6% loss=0.038986 lr=7.863203e-08 elapsed=35:10:55 avg_step=1.34s recent_step=1.24s eta=01:59:47 +2026-08-26 15:06:37 | INFO | train rank=0 | step=94640/100000 progress=94.6% loss=0.023880 lr=7.834022e-08 elapsed=35:11:06 avg_step=1.34s recent_step=1.15s eta=01:59:33 +2026-08-26 15:06:48 | INFO | train rank=0 | step=94650/100000 progress=94.7% loss=0.037109 lr=7.804894e-08 elapsed=35:11:18 avg_step=1.34s recent_step=1.19s eta=01:59:20 +2026-08-26 15:07:00 | INFO | train rank=0 | step=94660/100000 progress=94.7% loss=0.039864 lr=7.775819e-08 elapsed=35:11:30 avg_step=1.34s recent_step=1.20s eta=01:59:06 +2026-08-26 15:07:12 | INFO | train rank=0 | step=94670/100000 progress=94.7% loss=0.035370 lr=7.746799e-08 elapsed=35:11:42 avg_step=1.34s recent_step=1.12s eta=01:58:53 +2026-08-26 15:07:24 | INFO | train rank=0 | step=94680/100000 progress=94.7% loss=0.037518 lr=7.717833e-08 elapsed=35:11:54 avg_step=1.34s recent_step=1.24s eta=01:58:39 +2026-08-26 15:07:36 | INFO | train rank=0 | step=94690/100000 progress=94.7% loss=0.074097 lr=7.688920e-08 elapsed=35:12:06 avg_step=1.34s recent_step=1.18s eta=01:58:26 +2026-08-26 15:07:47 | INFO | train rank=0 | step=94700/100000 progress=94.7% loss=0.049271 lr=7.660061e-08 elapsed=35:12:17 avg_step=1.34s recent_step=1.09s eta=01:58:12 +2026-08-26 15:08:07 | INFO | train rank=0 | step=94710/100000 progress=94.7% loss=0.046783 lr=7.631256e-08 elapsed=35:12:37 avg_step=1.34s recent_step=2.00s eta=01:57:59 +2026-08-26 15:08:19 | INFO | train rank=0 | step=94720/100000 progress=94.7% loss=0.042542 lr=7.602505e-08 elapsed=35:12:49 avg_step=1.34s recent_step=1.20s eta=01:57:46 +2026-08-26 15:08:34 | INFO | train rank=0 | step=94730/100000 progress=94.7% loss=0.030106 lr=7.573808e-08 elapsed=35:13:04 avg_step=1.34s recent_step=1.52s eta=01:57:33 +2026-08-26 15:08:46 | INFO | train rank=0 | step=94740/100000 progress=94.7% loss=0.062622 lr=7.545165e-08 elapsed=35:13:16 avg_step=1.34s recent_step=1.21s eta=01:57:19 +2026-08-26 15:09:04 | INFO | train rank=0 | step=94750/100000 progress=94.8% loss=0.044403 lr=7.516575e-08 elapsed=35:13:34 avg_step=1.34s recent_step=1.82s eta=01:57:06 +2026-08-26 15:09:16 | INFO | train rank=0 | step=94760/100000 progress=94.8% loss=0.065735 lr=7.488039e-08 elapsed=35:13:46 avg_step=1.34s recent_step=1.20s eta=01:56:53 +2026-08-26 15:09:30 | INFO | train rank=0 | step=94770/100000 progress=94.8% loss=0.021484 lr=7.459558e-08 elapsed=35:14:00 avg_step=1.34s recent_step=1.39s eta=01:56:39 +2026-08-26 15:09:47 | INFO | train rank=0 | step=94780/100000 progress=94.8% loss=0.056244 lr=7.431130e-08 elapsed=35:14:16 avg_step=1.34s recent_step=1.63s eta=01:56:26 +2026-08-26 15:09:59 | INFO | train rank=0 | step=94790/100000 progress=94.8% loss=0.046021 lr=7.402756e-08 elapsed=35:14:29 avg_step=1.34s recent_step=1.23s eta=01:56:13 +2026-08-26 15:10:10 | INFO | train rank=0 | step=94800/100000 progress=94.8% loss=0.041618 lr=7.374435e-08 elapsed=35:14:40 avg_step=1.34s recent_step=1.13s eta=01:55:59 +2026-08-26 15:10:24 | INFO | train rank=0 | step=94810/100000 progress=94.8% loss=0.045364 lr=7.346169e-08 elapsed=35:14:54 avg_step=1.34s recent_step=1.35s eta=01:55:46 +2026-08-26 15:10:41 | INFO | train rank=0 | step=94820/100000 progress=94.8% loss=0.044586 lr=7.317957e-08 elapsed=35:15:11 avg_step=1.34s recent_step=1.70s eta=01:55:33 +2026-08-26 15:10:54 | INFO | train rank=0 | step=94830/100000 progress=94.8% loss=0.052399 lr=7.289798e-08 elapsed=35:15:24 avg_step=1.34s recent_step=1.33s eta=01:55:19 +2026-08-26 15:11:06 | INFO | train rank=0 | step=94840/100000 progress=94.8% loss=0.061249 lr=7.261693e-08 elapsed=35:15:36 avg_step=1.34s recent_step=1.19s eta=01:55:06 +2026-08-26 15:11:18 | INFO | train rank=0 | step=94850/100000 progress=94.8% loss=0.033218 lr=7.233642e-08 elapsed=35:15:48 avg_step=1.34s recent_step=1.19s eta=01:54:52 +2026-08-26 15:11:31 | INFO | train rank=0 | step=94860/100000 progress=94.9% loss=0.041580 lr=7.205645e-08 elapsed=35:16:01 avg_step=1.34s recent_step=1.30s eta=01:54:39 +2026-08-26 15:11:43 | INFO | train rank=0 | step=94870/100000 progress=94.9% loss=0.050293 lr=7.177702e-08 elapsed=35:16:13 avg_step=1.34s recent_step=1.22s eta=01:54:25 +2026-08-26 15:11:57 | INFO | train rank=0 | step=94880/100000 progress=94.9% loss=0.039314 lr=7.149813e-08 elapsed=35:16:27 avg_step=1.34s recent_step=1.39s eta=01:54:12 +2026-08-26 15:12:12 | INFO | train rank=0 | step=94890/100000 progress=94.9% loss=0.042770 lr=7.121978e-08 elapsed=35:16:42 avg_step=1.34s recent_step=1.50s eta=01:53:59 +2026-08-26 15:12:23 | INFO | train rank=0 | step=94900/100000 progress=94.9% loss=0.046844 lr=7.094197e-08 elapsed=35:16:53 avg_step=1.34s recent_step=1.17s eta=01:53:45 +2026-08-26 15:12:38 | INFO | train rank=0 | step=94910/100000 progress=94.9% loss=0.031952 lr=7.066469e-08 elapsed=35:17:08 avg_step=1.34s recent_step=1.43s eta=01:53:32 +2026-08-26 15:12:56 | INFO | train rank=0 | step=94920/100000 progress=94.9% loss=0.048431 lr=7.038796e-08 elapsed=35:17:26 avg_step=1.34s recent_step=1.82s eta=01:53:19 +2026-08-26 15:13:10 | INFO | train rank=0 | step=94930/100000 progress=94.9% loss=0.035416 lr=7.011176e-08 elapsed=35:17:40 avg_step=1.34s recent_step=1.42s eta=01:53:06 +2026-08-26 15:13:21 | INFO | train rank=0 | step=94940/100000 progress=94.9% loss=0.047852 lr=6.983610e-08 elapsed=35:17:51 avg_step=1.34s recent_step=1.15s eta=01:52:52 +2026-08-26 15:13:33 | INFO | train rank=0 | step=94950/100000 progress=95.0% loss=0.057800 lr=6.956098e-08 elapsed=35:18:03 avg_step=1.34s recent_step=1.18s eta=01:52:39 +2026-08-26 15:13:47 | INFO | train rank=0 | step=94960/100000 progress=95.0% loss=0.035797 lr=6.928641e-08 elapsed=35:18:17 avg_step=1.34s recent_step=1.36s eta=01:52:25 +2026-08-26 15:14:00 | INFO | train rank=0 | step=94970/100000 progress=95.0% loss=0.043785 lr=6.901237e-08 elapsed=35:18:29 avg_step=1.34s recent_step=1.28s eta=01:52:12 +2026-08-26 15:14:15 | INFO | train rank=0 | step=94980/100000 progress=95.0% loss=0.050842 lr=6.873887e-08 elapsed=35:18:45 avg_step=1.34s recent_step=1.54s eta=01:51:58 +2026-08-26 15:14:29 | INFO | train rank=0 | step=94990/100000 progress=95.0% loss=0.048706 lr=6.846590e-08 elapsed=35:18:59 avg_step=1.34s recent_step=1.43s eta=01:51:45 +2026-08-26 15:14:41 | INFO | train rank=0 | step=95000/100000 progress=95.0% loss=0.042145 lr=6.819348e-08 elapsed=35:19:11 avg_step=1.34s recent_step=1.16s eta=01:51:32 +2026-08-26 15:14:55 | INFO | train rank=0 | step=95010/100000 progress=95.0% loss=0.042343 lr=6.792160e-08 elapsed=35:19:24 avg_step=1.34s recent_step=1.37s eta=01:51:18 +2026-08-26 15:15:09 | INFO | train rank=0 | step=95020/100000 progress=95.0% loss=0.058289 lr=6.765026e-08 elapsed=35:19:38 avg_step=1.34s recent_step=1.40s eta=01:51:05 +2026-08-26 15:15:21 | INFO | train rank=0 | step=95030/100000 progress=95.0% loss=0.035484 lr=6.737945e-08 elapsed=35:19:50 avg_step=1.34s recent_step=1.19s eta=01:50:51 +2026-08-26 15:15:35 | INFO | train rank=0 | step=95040/100000 progress=95.0% loss=0.055695 lr=6.710919e-08 elapsed=35:20:05 avg_step=1.34s recent_step=1.48s eta=01:50:38 +2026-08-26 15:15:47 | INFO | train rank=0 | step=95050/100000 progress=95.0% loss=0.051254 lr=6.683946e-08 elapsed=35:20:17 avg_step=1.34s recent_step=1.15s eta=01:50:25 +2026-08-26 15:15:58 | INFO | train rank=0 | step=95060/100000 progress=95.1% loss=0.030327 lr=6.657028e-08 elapsed=35:20:28 avg_step=1.34s recent_step=1.16s eta=01:50:11 +2026-08-26 15:16:14 | INFO | train rank=0 | step=95070/100000 progress=95.1% loss=0.046425 lr=6.630163e-08 elapsed=35:20:44 avg_step=1.34s recent_step=1.53s eta=01:49:58 +2026-08-26 15:16:27 | INFO | train rank=0 | step=95080/100000 progress=95.1% loss=0.044907 lr=6.603353e-08 elapsed=35:20:57 avg_step=1.34s recent_step=1.34s eta=01:49:45 +2026-08-26 15:16:39 | INFO | train rank=0 | step=95090/100000 progress=95.1% loss=0.054077 lr=6.576596e-08 elapsed=35:21:09 avg_step=1.34s recent_step=1.20s eta=01:49:31 +2026-08-26 15:16:54 | INFO | train rank=0 | step=95100/100000 progress=95.1% loss=0.061630 lr=6.549893e-08 elapsed=35:21:24 avg_step=1.34s recent_step=1.54s eta=01:49:18 +2026-08-26 15:17:07 | INFO | train rank=0 | step=95110/100000 progress=95.1% loss=0.042572 lr=6.523245e-08 elapsed=35:21:36 avg_step=1.34s recent_step=1.21s eta=01:49:04 +2026-08-26 15:17:18 | INFO | train rank=0 | step=95120/100000 progress=95.1% loss=0.068695 lr=6.496650e-08 elapsed=35:21:48 avg_step=1.34s recent_step=1.19s eta=01:48:51 +2026-08-26 15:17:35 | INFO | train rank=0 | step=95130/100000 progress=95.1% loss=0.049965 lr=6.470109e-08 elapsed=35:22:05 avg_step=1.34s recent_step=1.62s eta=01:48:38 +2026-08-26 15:17:46 | INFO | train rank=0 | step=95140/100000 progress=95.1% loss=0.039825 lr=6.443622e-08 elapsed=35:22:16 avg_step=1.34s recent_step=1.18s eta=01:48:24 +2026-08-26 15:18:01 | INFO | train rank=0 | step=95150/100000 progress=95.2% loss=0.062714 lr=6.417189e-08 elapsed=35:22:31 avg_step=1.34s recent_step=1.47s eta=01:48:11 +2026-08-26 15:18:13 | INFO | train rank=0 | step=95160/100000 progress=95.2% loss=0.039848 lr=6.390810e-08 elapsed=35:22:43 avg_step=1.34s recent_step=1.19s eta=01:47:57 +2026-08-26 15:18:28 | INFO | train rank=0 | step=95170/100000 progress=95.2% loss=0.050522 lr=6.364485e-08 elapsed=35:22:58 avg_step=1.34s recent_step=1.53s eta=01:47:44 +2026-08-26 15:18:44 | INFO | train rank=0 | step=95180/100000 progress=95.2% loss=0.024078 lr=6.338214e-08 elapsed=35:23:14 avg_step=1.34s recent_step=1.60s eta=01:47:31 +2026-08-26 15:18:56 | INFO | train rank=0 | step=95190/100000 progress=95.2% loss=0.039597 lr=6.311998e-08 elapsed=35:23:26 avg_step=1.34s recent_step=1.16s eta=01:47:17 +2026-08-26 15:19:10 | INFO | train rank=0 | step=95200/100000 progress=95.2% loss=0.056885 lr=6.285835e-08 elapsed=35:23:40 avg_step=1.34s recent_step=1.43s eta=01:47:04 +2026-08-26 15:19:23 | INFO | train rank=0 | step=95210/100000 progress=95.2% loss=0.037247 lr=6.259726e-08 elapsed=35:23:53 avg_step=1.34s recent_step=1.30s eta=01:46:51 +2026-08-26 15:19:38 | INFO | train rank=0 | step=95220/100000 progress=95.2% loss=0.039490 lr=6.233671e-08 elapsed=35:24:08 avg_step=1.34s recent_step=1.49s eta=01:46:37 +2026-08-26 15:19:52 | INFO | train rank=0 | step=95230/100000 progress=95.2% loss=0.027802 lr=6.207670e-08 elapsed=35:24:22 avg_step=1.34s recent_step=1.39s eta=01:46:24 +2026-08-26 15:20:03 | INFO | train rank=0 | step=95240/100000 progress=95.2% loss=0.055084 lr=6.181723e-08 elapsed=35:24:33 avg_step=1.34s recent_step=1.14s eta=01:46:11 +2026-08-26 15:20:18 | INFO | train rank=0 | step=95250/100000 progress=95.2% loss=0.060394 lr=6.155830e-08 elapsed=35:24:48 avg_step=1.34s recent_step=1.50s eta=01:45:57 +2026-08-26 15:20:35 | INFO | train rank=0 | step=95260/100000 progress=95.3% loss=0.048294 lr=6.129991e-08 elapsed=35:25:05 avg_step=1.34s recent_step=1.66s eta=01:45:44 +2026-08-26 15:20:48 | INFO | train rank=0 | step=95270/100000 progress=95.3% loss=0.051300 lr=6.104206e-08 elapsed=35:25:18 avg_step=1.34s recent_step=1.33s eta=01:45:31 +2026-08-26 15:21:00 | INFO | train rank=0 | step=95280/100000 progress=95.3% loss=0.043716 lr=6.078475e-08 elapsed=35:25:29 avg_step=1.34s recent_step=1.14s eta=01:45:17 +2026-08-26 15:21:15 | INFO | train rank=0 | step=95290/100000 progress=95.3% loss=0.027058 lr=6.052798e-08 elapsed=35:25:45 avg_step=1.34s recent_step=1.52s eta=01:45:04 +2026-08-26 15:21:27 | INFO | train rank=0 | step=95300/100000 progress=95.3% loss=0.033951 lr=6.027175e-08 elapsed=35:25:57 avg_step=1.34s recent_step=1.22s eta=01:44:50 +2026-08-26 15:21:39 | INFO | train rank=0 | step=95310/100000 progress=95.3% loss=0.035294 lr=6.001606e-08 elapsed=35:26:09 avg_step=1.34s recent_step=1.26s eta=01:44:37 +2026-08-26 15:21:52 | INFO | train rank=0 | step=95320/100000 progress=95.3% loss=0.047371 lr=5.976091e-08 elapsed=35:26:22 avg_step=1.34s recent_step=1.28s eta=01:44:24 +2026-08-26 15:22:08 | INFO | train rank=0 | step=95330/100000 progress=95.3% loss=0.027557 lr=5.950630e-08 elapsed=35:26:38 avg_step=1.34s recent_step=1.54s eta=01:44:10 +2026-08-26 15:22:20 | INFO | train rank=0 | step=95340/100000 progress=95.3% loss=0.033508 lr=5.925224e-08 elapsed=35:26:50 avg_step=1.34s recent_step=1.23s eta=01:43:57 +2026-08-26 15:22:35 | INFO | train rank=0 | step=95350/100000 progress=95.3% loss=0.041016 lr=5.899871e-08 elapsed=35:27:05 avg_step=1.34s recent_step=1.51s eta=01:43:43 +2026-08-26 15:22:50 | INFO | train rank=0 | step=95360/100000 progress=95.4% loss=0.040649 lr=5.874572e-08 elapsed=35:27:20 avg_step=1.34s recent_step=1.46s eta=01:43:30 +2026-08-26 15:23:04 | INFO | train rank=0 | step=95370/100000 progress=95.4% loss=0.026093 lr=5.849327e-08 elapsed=35:27:34 avg_step=1.34s recent_step=1.44s eta=01:43:17 +2026-08-26 15:23:18 | INFO | train rank=0 | step=95380/100000 progress=95.4% loss=0.037529 lr=5.824137e-08 elapsed=35:27:48 avg_step=1.34s recent_step=1.37s eta=01:43:03 +2026-08-26 15:23:29 | INFO | train rank=0 | step=95390/100000 progress=95.4% loss=0.034775 lr=5.799000e-08 elapsed=35:27:59 avg_step=1.34s recent_step=1.15s eta=01:42:50 +2026-08-26 15:23:44 | INFO | train rank=0 | step=95400/100000 progress=95.4% loss=0.051521 lr=5.773917e-08 elapsed=35:28:14 avg_step=1.34s recent_step=1.44s eta=01:42:37 +2026-08-26 15:23:59 | INFO | train rank=0 | step=95410/100000 progress=95.4% loss=0.054581 lr=5.748889e-08 elapsed=35:28:29 avg_step=1.34s recent_step=1.50s eta=01:42:23 +2026-08-26 15:24:12 | INFO | train rank=0 | step=95420/100000 progress=95.4% loss=0.029419 lr=5.723914e-08 elapsed=35:28:42 avg_step=1.34s recent_step=1.37s eta=01:42:10 +2026-08-26 15:24:24 | INFO | train rank=0 | step=95430/100000 progress=95.4% loss=0.060364 lr=5.698994e-08 elapsed=35:28:54 avg_step=1.34s recent_step=1.16s eta=01:41:57 +2026-08-26 15:24:39 | INFO | train rank=0 | step=95440/100000 progress=95.4% loss=0.040955 lr=5.674128e-08 elapsed=35:29:09 avg_step=1.34s recent_step=1.46s eta=01:41:43 +2026-08-26 15:24:51 | INFO | train rank=0 | step=95450/100000 progress=95.5% loss=0.053146 lr=5.649315e-08 elapsed=35:29:21 avg_step=1.34s recent_step=1.25s eta=01:41:30 +2026-08-26 15:25:06 | INFO | train rank=0 | step=95460/100000 progress=95.5% loss=0.047119 lr=5.624557e-08 elapsed=35:29:36 avg_step=1.34s recent_step=1.46s eta=01:41:16 +2026-08-26 15:25:18 | INFO | train rank=0 | step=95470/100000 progress=95.5% loss=0.050018 lr=5.599853e-08 elapsed=35:29:48 avg_step=1.34s recent_step=1.20s eta=01:41:03 +2026-08-26 15:25:29 | INFO | train rank=0 | step=95480/100000 progress=95.5% loss=0.053802 lr=5.575203e-08 elapsed=35:29:59 avg_step=1.34s recent_step=1.13s eta=01:40:49 +2026-08-26 15:25:42 | INFO | train rank=0 | step=95490/100000 progress=95.5% loss=0.053398 lr=5.550607e-08 elapsed=35:30:12 avg_step=1.34s recent_step=1.35s eta=01:40:36 +2026-08-26 15:25:54 | INFO | train rank=0 | step=95500/100000 progress=95.5% loss=0.080811 lr=5.526065e-08 elapsed=35:30:24 avg_step=1.34s recent_step=1.14s eta=01:40:23 +2026-08-26 15:26:06 | INFO | train rank=0 | step=95510/100000 progress=95.5% loss=0.053650 lr=5.501577e-08 elapsed=35:30:36 avg_step=1.34s recent_step=1.19s eta=01:40:09 +2026-08-26 15:26:26 | INFO | train rank=0 | step=95520/100000 progress=95.5% loss=0.041473 lr=5.477143e-08 elapsed=35:30:56 avg_step=1.34s recent_step=2.00s eta=01:39:56 +2026-08-26 15:26:37 | INFO | train rank=0 | step=95530/100000 progress=95.5% loss=0.036339 lr=5.452763e-08 elapsed=35:31:07 avg_step=1.34s recent_step=1.15s eta=01:39:43 +2026-08-26 15:26:55 | INFO | train rank=0 | step=95540/100000 progress=95.5% loss=0.071503 lr=5.428438e-08 elapsed=35:31:24 avg_step=1.34s recent_step=1.72s eta=01:39:29 +2026-08-26 15:27:06 | INFO | train rank=0 | step=95550/100000 progress=95.5% loss=0.046524 lr=5.404166e-08 elapsed=35:31:36 avg_step=1.34s recent_step=1.17s eta=01:39:16 +2026-08-26 15:27:18 | INFO | train rank=0 | step=95560/100000 progress=95.6% loss=0.074829 lr=5.379948e-08 elapsed=35:31:48 avg_step=1.34s recent_step=1.18s eta=01:39:03 +2026-08-26 15:27:35 | INFO | train rank=0 | step=95570/100000 progress=95.6% loss=0.053101 lr=5.355785e-08 elapsed=35:32:05 avg_step=1.34s recent_step=1.71s eta=01:38:49 +2026-08-26 15:27:47 | INFO | train rank=0 | step=95580/100000 progress=95.6% loss=0.038597 lr=5.331676e-08 elapsed=35:32:17 avg_step=1.34s recent_step=1.20s eta=01:38:36 +2026-08-26 15:28:00 | INFO | train rank=0 | step=95590/100000 progress=95.6% loss=0.042534 lr=5.307621e-08 elapsed=35:32:30 avg_step=1.34s recent_step=1.30s eta=01:38:22 +2026-08-26 15:28:13 | INFO | train rank=0 | step=95600/100000 progress=95.6% loss=0.025017 lr=5.283619e-08 elapsed=35:32:43 avg_step=1.34s recent_step=1.29s eta=01:38:09 +2026-08-26 15:28:25 | INFO | train rank=0 | step=95610/100000 progress=95.6% loss=0.070007 lr=5.259672e-08 elapsed=35:32:55 avg_step=1.34s recent_step=1.17s eta=01:37:56 +2026-08-26 15:28:40 | INFO | train rank=0 | step=95620/100000 progress=95.6% loss=0.045456 lr=5.235780e-08 elapsed=35:33:10 avg_step=1.34s recent_step=1.56s eta=01:37:42 +2026-08-26 15:28:52 | INFO | train rank=0 | step=95630/100000 progress=95.6% loss=0.063751 lr=5.211941e-08 elapsed=35:33:22 avg_step=1.34s recent_step=1.21s eta=01:37:29 +2026-08-26 15:29:06 | INFO | train rank=0 | step=95640/100000 progress=95.6% loss=0.046539 lr=5.188156e-08 elapsed=35:33:36 avg_step=1.34s recent_step=1.33s eta=01:37:15 +2026-08-26 15:29:20 | INFO | train rank=0 | step=95650/100000 progress=95.7% loss=0.088745 lr=5.164425e-08 elapsed=35:33:50 avg_step=1.34s recent_step=1.44s eta=01:37:02 +2026-08-26 15:29:37 | INFO | train rank=0 | step=95660/100000 progress=95.7% loss=0.033127 lr=5.140749e-08 elapsed=35:34:07 avg_step=1.34s recent_step=1.69s eta=01:36:49 +2026-08-26 15:29:49 | INFO | train rank=0 | step=95670/100000 progress=95.7% loss=0.049744 lr=5.117127e-08 elapsed=35:34:18 avg_step=1.34s recent_step=1.15s eta=01:36:35 +2026-08-26 15:30:05 | INFO | train rank=0 | step=95680/100000 progress=95.7% loss=0.052917 lr=5.093558e-08 elapsed=35:34:35 avg_step=1.34s recent_step=1.69s eta=01:36:22 +2026-08-26 15:30:16 | INFO | train rank=0 | step=95690/100000 progress=95.7% loss=0.121704 lr=5.070044e-08 elapsed=35:34:46 avg_step=1.34s recent_step=1.11s eta=01:36:09 +2026-08-26 15:30:29 | INFO | train rank=0 | step=95700/100000 progress=95.7% loss=0.049637 lr=5.046584e-08 elapsed=35:34:59 avg_step=1.34s recent_step=1.22s eta=01:35:55 +2026-08-26 15:30:42 | INFO | train rank=0 | step=95710/100000 progress=95.7% loss=0.048965 lr=5.023178e-08 elapsed=35:35:12 avg_step=1.34s recent_step=1.34s eta=01:35:42 +2026-08-26 15:30:53 | INFO | train rank=0 | step=95720/100000 progress=95.7% loss=0.065186 lr=4.999827e-08 elapsed=35:35:23 avg_step=1.34s recent_step=1.08s eta=01:35:28 +2026-08-26 15:31:05 | INFO | train rank=0 | step=95730/100000 progress=95.7% loss=0.040573 lr=4.976529e-08 elapsed=35:35:35 avg_step=1.34s recent_step=1.24s eta=01:35:15 +2026-08-26 15:31:22 | INFO | train rank=0 | step=95740/100000 progress=95.7% loss=0.028549 lr=4.953286e-08 elapsed=35:35:52 avg_step=1.34s recent_step=1.64s eta=01:35:02 +2026-08-26 15:31:37 | INFO | train rank=0 | step=95750/100000 progress=95.8% loss=0.031952 lr=4.930096e-08 elapsed=35:36:06 avg_step=1.34s recent_step=1.49s eta=01:34:48 +2026-08-26 15:31:48 | INFO | train rank=0 | step=95760/100000 progress=95.8% loss=0.090698 lr=4.906961e-08 elapsed=35:36:18 avg_step=1.34s recent_step=1.19s eta=01:34:35 +2026-08-26 15:32:04 | INFO | train rank=0 | step=95770/100000 progress=95.8% loss=0.050537 lr=4.883880e-08 elapsed=35:36:34 avg_step=1.34s recent_step=1.61s eta=01:34:22 +2026-08-26 15:32:17 | INFO | train rank=0 | step=95780/100000 progress=95.8% loss=0.035858 lr=4.860853e-08 elapsed=35:36:47 avg_step=1.34s recent_step=1.30s eta=01:34:08 +2026-08-26 15:32:29 | INFO | train rank=0 | step=95790/100000 progress=95.8% loss=0.043579 lr=4.837880e-08 elapsed=35:36:59 avg_step=1.34s recent_step=1.19s eta=01:33:55 +2026-08-26 15:32:40 | INFO | train rank=0 | step=95800/100000 progress=95.8% loss=0.025269 lr=4.814962e-08 elapsed=35:37:10 avg_step=1.34s recent_step=1.11s eta=01:33:41 +2026-08-26 15:32:54 | INFO | train rank=0 | step=95810/100000 progress=95.8% loss=0.050552 lr=4.792097e-08 elapsed=35:37:24 avg_step=1.34s recent_step=1.32s eta=01:33:28 +2026-08-26 15:33:05 | INFO | train rank=0 | step=95820/100000 progress=95.8% loss=0.062317 lr=4.769287e-08 elapsed=35:37:35 avg_step=1.34s recent_step=1.17s eta=01:33:14 +2026-08-26 15:33:20 | INFO | train rank=0 | step=95830/100000 progress=95.8% loss=0.055847 lr=4.746531e-08 elapsed=35:37:50 avg_step=1.34s recent_step=1.51s eta=01:33:01 +2026-08-26 15:33:36 | INFO | train rank=0 | step=95840/100000 progress=95.8% loss=0.043457 lr=4.723829e-08 elapsed=35:38:06 avg_step=1.34s recent_step=1.56s eta=01:32:48 +2026-08-26 15:33:50 | INFO | train rank=0 | step=95850/100000 progress=95.9% loss=0.071198 lr=4.701181e-08 elapsed=35:38:20 avg_step=1.34s recent_step=1.44s eta=01:32:35 +2026-08-26 15:34:04 | INFO | train rank=0 | step=95860/100000 progress=95.9% loss=0.028934 lr=4.678588e-08 elapsed=35:38:34 avg_step=1.34s recent_step=1.35s eta=01:32:21 +2026-08-26 15:34:17 | INFO | train rank=0 | step=95870/100000 progress=95.9% loss=0.060028 lr=4.656048e-08 elapsed=35:38:47 avg_step=1.34s recent_step=1.29s eta=01:32:08 +2026-08-26 15:34:28 | INFO | train rank=0 | step=95880/100000 progress=95.9% loss=0.043900 lr=4.633563e-08 elapsed=35:38:58 avg_step=1.34s recent_step=1.11s eta=01:31:54 +2026-08-26 15:34:45 | INFO | train rank=0 | step=95890/100000 progress=95.9% loss=0.044327 lr=4.611132e-08 elapsed=35:39:15 avg_step=1.34s recent_step=1.68s eta=01:31:41 +2026-08-26 15:34:56 | INFO | train rank=0 | step=95900/100000 progress=95.9% loss=0.034790 lr=4.588755e-08 elapsed=35:39:26 avg_step=1.34s recent_step=1.12s eta=01:31:28 +2026-08-26 15:35:08 | INFO | train rank=0 | step=95910/100000 progress=95.9% loss=0.086456 lr=4.566432e-08 elapsed=35:39:38 avg_step=1.34s recent_step=1.20s eta=01:31:14 +2026-08-26 15:35:20 | INFO | train rank=0 | step=95920/100000 progress=95.9% loss=0.065857 lr=4.544163e-08 elapsed=35:39:50 avg_step=1.34s recent_step=1.18s eta=01:31:01 +2026-08-26 15:35:38 | INFO | train rank=0 | step=95930/100000 progress=95.9% loss=0.031395 lr=4.521949e-08 elapsed=35:40:08 avg_step=1.34s recent_step=1.80s eta=01:30:47 +2026-08-26 15:35:50 | INFO | train rank=0 | step=95940/100000 progress=95.9% loss=0.037231 lr=4.499789e-08 elapsed=35:40:20 avg_step=1.34s recent_step=1.26s eta=01:30:34 +2026-08-26 15:36:05 | INFO | train rank=0 | step=95950/100000 progress=96.0% loss=0.057510 lr=4.477683e-08 elapsed=35:40:35 avg_step=1.34s recent_step=1.50s eta=01:30:21 +2026-08-26 15:36:17 | INFO | train rank=0 | step=95960/100000 progress=96.0% loss=0.042572 lr=4.455631e-08 elapsed=35:40:47 avg_step=1.34s recent_step=1.13s eta=01:30:07 +2026-08-26 15:36:29 | INFO | train rank=0 | step=95970/100000 progress=96.0% loss=0.061584 lr=4.433633e-08 elapsed=35:40:59 avg_step=1.34s recent_step=1.21s eta=01:29:54 +2026-08-26 15:36:40 | INFO | train rank=0 | step=95980/100000 progress=96.0% loss=0.032749 lr=4.411690e-08 elapsed=35:41:10 avg_step=1.34s recent_step=1.11s eta=01:29:40 +2026-08-26 15:36:55 | INFO | train rank=0 | step=95990/100000 progress=96.0% loss=0.041458 lr=4.389800e-08 elapsed=35:41:25 avg_step=1.34s recent_step=1.55s eta=01:29:27 +2026-08-26 15:37:07 | INFO | train rank=0 | step=96000/100000 progress=96.0% loss=0.046326 lr=4.367965e-08 elapsed=35:41:37 avg_step=1.34s recent_step=1.14s eta=01:29:14 +2026-08-26 15:37:20 | INFO | train rank=0 | step=96010/100000 progress=96.0% loss=0.041687 lr=4.346184e-08 elapsed=35:41:50 avg_step=1.34s recent_step=1.37s eta=01:29:00 +2026-08-26 15:37:33 | INFO | train rank=0 | step=96020/100000 progress=96.0% loss=0.046463 lr=4.324458e-08 elapsed=35:42:03 avg_step=1.34s recent_step=1.24s eta=01:28:47 +2026-08-26 15:37:47 | INFO | train rank=0 | step=96030/100000 progress=96.0% loss=0.036865 lr=4.302785e-08 elapsed=35:42:17 avg_step=1.34s recent_step=1.42s eta=01:28:33 +2026-08-26 15:38:01 | INFO | train rank=0 | step=96040/100000 progress=96.0% loss=0.049423 lr=4.281167e-08 elapsed=35:42:31 avg_step=1.34s recent_step=1.37s eta=01:28:20 +2026-08-26 15:38:19 | INFO | train rank=0 | step=96050/100000 progress=96.0% loss=0.050980 lr=4.259603e-08 elapsed=35:42:49 avg_step=1.34s recent_step=1.85s eta=01:28:07 +2026-08-26 15:38:32 | INFO | train rank=0 | step=96060/100000 progress=96.1% loss=0.048767 lr=4.238093e-08 elapsed=35:43:02 avg_step=1.34s recent_step=1.28s eta=01:27:53 +2026-08-26 15:38:46 | INFO | train rank=0 | step=96070/100000 progress=96.1% loss=0.046722 lr=4.216638e-08 elapsed=35:43:16 avg_step=1.34s recent_step=1.40s eta=01:27:40 +2026-08-26 15:39:02 | INFO | train rank=0 | step=96080/100000 progress=96.1% loss=0.047615 lr=4.195236e-08 elapsed=35:43:32 avg_step=1.34s recent_step=1.61s eta=01:27:27 +2026-08-26 15:39:18 | INFO | train rank=0 | step=96090/100000 progress=96.1% loss=0.040894 lr=4.173889e-08 elapsed=35:43:48 avg_step=1.34s recent_step=1.62s eta=01:27:14 +2026-08-26 15:39:30 | INFO | train rank=0 | step=96100/100000 progress=96.1% loss=0.028728 lr=4.152596e-08 elapsed=35:44:00 avg_step=1.34s recent_step=1.15s eta=01:27:00 +2026-08-26 15:39:44 | INFO | train rank=0 | step=96110/100000 progress=96.1% loss=0.037254 lr=4.131357e-08 elapsed=35:44:14 avg_step=1.34s recent_step=1.39s eta=01:26:47 +2026-08-26 15:39:57 | INFO | train rank=0 | step=96120/100000 progress=96.1% loss=0.037064 lr=4.110173e-08 elapsed=35:44:27 avg_step=1.34s recent_step=1.34s eta=01:26:33 +2026-08-26 15:40:10 | INFO | train rank=0 | step=96130/100000 progress=96.1% loss=0.036484 lr=4.089043e-08 elapsed=35:44:40 avg_step=1.34s recent_step=1.32s eta=01:26:20 +2026-08-26 15:40:22 | INFO | train rank=0 | step=96140/100000 progress=96.1% loss=0.042419 lr=4.067967e-08 elapsed=35:44:52 avg_step=1.34s recent_step=1.19s eta=01:26:06 +2026-08-26 15:40:37 | INFO | train rank=0 | step=96150/100000 progress=96.2% loss=0.026585 lr=4.046945e-08 elapsed=35:45:07 avg_step=1.34s recent_step=1.49s eta=01:25:53 +2026-08-26 15:40:52 | INFO | train rank=0 | step=96160/100000 progress=96.2% loss=0.040863 lr=4.025977e-08 elapsed=35:45:22 avg_step=1.34s recent_step=1.50s eta=01:25:40 +2026-08-26 15:41:09 | INFO | train rank=0 | step=96170/100000 progress=96.2% loss=0.102600 lr=4.005064e-08 elapsed=35:45:39 avg_step=1.34s recent_step=1.68s eta=01:25:27 +2026-08-26 15:41:21 | INFO | train rank=0 | step=96180/100000 progress=96.2% loss=0.030312 lr=3.984205e-08 elapsed=35:45:51 avg_step=1.34s recent_step=1.18s eta=01:25:13 +2026-08-26 15:41:35 | INFO | train rank=0 | step=96190/100000 progress=96.2% loss=0.041641 lr=3.963400e-08 elapsed=35:46:05 avg_step=1.34s recent_step=1.46s eta=01:25:00 +2026-08-26 15:41:51 | INFO | train rank=0 | step=96200/100000 progress=96.2% loss=0.047119 lr=3.942649e-08 elapsed=35:46:21 avg_step=1.34s recent_step=1.56s eta=01:24:47 +2026-08-26 15:42:02 | INFO | train rank=0 | step=96210/100000 progress=96.2% loss=0.031837 lr=3.921953e-08 elapsed=35:46:32 avg_step=1.34s recent_step=1.14s eta=01:24:33 +2026-08-26 15:42:16 | INFO | train rank=0 | step=96220/100000 progress=96.2% loss=0.046417 lr=3.901311e-08 elapsed=35:46:46 avg_step=1.34s recent_step=1.31s eta=01:24:20 +2026-08-26 15:42:28 | INFO | train rank=0 | step=96230/100000 progress=96.2% loss=0.032463 lr=3.880723e-08 elapsed=35:46:58 avg_step=1.34s recent_step=1.22s eta=01:24:06 +2026-08-26 15:42:44 | INFO | train rank=0 | step=96240/100000 progress=96.2% loss=0.075226 lr=3.860189e-08 elapsed=35:47:14 avg_step=1.34s recent_step=1.59s eta=01:23:53 +2026-08-26 15:42:55 | INFO | train rank=0 | step=96250/100000 progress=96.2% loss=0.042107 lr=3.839710e-08 elapsed=35:47:25 avg_step=1.34s recent_step=1.12s eta=01:23:39 +2026-08-26 15:43:10 | INFO | train rank=0 | step=96260/100000 progress=96.3% loss=0.063507 lr=3.819285e-08 elapsed=35:47:40 avg_step=1.34s recent_step=1.50s eta=01:23:26 +2026-08-26 15:43:22 | INFO | train rank=0 | step=96270/100000 progress=96.3% loss=0.072571 lr=3.798914e-08 elapsed=35:47:51 avg_step=1.34s recent_step=1.15s eta=01:23:13 +2026-08-26 15:43:39 | INFO | train rank=0 | step=96280/100000 progress=96.3% loss=0.031662 lr=3.778598e-08 elapsed=35:48:09 avg_step=1.34s recent_step=1.73s eta=01:22:59 +2026-08-26 15:43:51 | INFO | train rank=0 | step=96290/100000 progress=96.3% loss=0.036613 lr=3.758335e-08 elapsed=35:48:21 avg_step=1.34s recent_step=1.21s eta=01:22:46 +2026-08-26 15:44:02 | INFO | train rank=0 | step=96300/100000 progress=96.3% loss=0.030304 lr=3.738127e-08 elapsed=35:48:32 avg_step=1.34s recent_step=1.12s eta=01:22:33 +2026-08-26 15:44:18 | INFO | train rank=0 | step=96310/100000 progress=96.3% loss=0.056854 lr=3.717974e-08 elapsed=35:48:48 avg_step=1.34s recent_step=1.59s eta=01:22:19 +2026-08-26 15:44:30 | INFO | train rank=0 | step=96320/100000 progress=96.3% loss=0.045921 lr=3.697874e-08 elapsed=35:49:00 avg_step=1.34s recent_step=1.20s eta=01:22:06 +2026-08-26 15:44:41 | INFO | train rank=0 | step=96330/100000 progress=96.3% loss=0.049450 lr=3.677829e-08 elapsed=35:49:11 avg_step=1.34s recent_step=1.12s eta=01:21:52 +2026-08-26 15:44:55 | INFO | train rank=0 | step=96340/100000 progress=96.3% loss=0.035034 lr=3.657838e-08 elapsed=35:49:25 avg_step=1.34s recent_step=1.42s eta=01:21:39 +2026-08-26 15:45:07 | INFO | train rank=0 | step=96350/100000 progress=96.4% loss=0.041458 lr=3.637901e-08 elapsed=35:49:37 avg_step=1.34s recent_step=1.16s eta=01:21:26 +2026-08-26 15:45:19 | INFO | train rank=0 | step=96360/100000 progress=96.4% loss=0.033195 lr=3.618019e-08 elapsed=35:49:49 avg_step=1.34s recent_step=1.24s eta=01:21:12 +2026-08-26 15:45:34 | INFO | train rank=0 | step=96370/100000 progress=96.4% loss=0.036324 lr=3.598191e-08 elapsed=35:50:04 avg_step=1.34s recent_step=1.43s eta=01:20:59 +2026-08-26 15:45:45 | INFO | train rank=0 | step=96380/100000 progress=96.4% loss=0.046997 lr=3.578417e-08 elapsed=35:50:15 avg_step=1.34s recent_step=1.15s eta=01:20:45 +2026-08-26 15:46:01 | INFO | train rank=0 | step=96390/100000 progress=96.4% loss=0.045311 lr=3.558698e-08 elapsed=35:50:30 avg_step=1.34s recent_step=1.52s eta=01:20:32 +2026-08-26 15:46:12 | INFO | train rank=0 | step=96400/100000 progress=96.4% loss=0.052460 lr=3.539033e-08 elapsed=35:50:42 avg_step=1.34s recent_step=1.15s eta=01:20:19 +2026-08-26 15:46:25 | INFO | train rank=0 | step=96410/100000 progress=96.4% loss=0.033180 lr=3.519422e-08 elapsed=35:50:55 avg_step=1.34s recent_step=1.28s eta=01:20:05 +2026-08-26 15:46:37 | INFO | train rank=0 | step=96420/100000 progress=96.4% loss=0.032166 lr=3.499865e-08 elapsed=35:51:07 avg_step=1.34s recent_step=1.26s eta=01:19:52 +2026-08-26 15:46:48 | INFO | train rank=0 | step=96430/100000 progress=96.4% loss=0.059723 lr=3.480363e-08 elapsed=35:51:18 avg_step=1.34s recent_step=1.10s eta=01:19:38 +2026-08-26 15:47:00 | INFO | train rank=0 | step=96440/100000 progress=96.4% loss=0.055939 lr=3.460915e-08 elapsed=35:51:30 avg_step=1.34s recent_step=1.18s eta=01:19:25 +2026-08-26 15:47:16 | INFO | train rank=0 | step=96450/100000 progress=96.5% loss=0.062271 lr=3.441521e-08 elapsed=35:51:46 avg_step=1.34s recent_step=1.58s eta=01:19:11 +2026-08-26 15:47:32 | INFO | train rank=0 | step=96460/100000 progress=96.5% loss=0.036407 lr=3.422181e-08 elapsed=35:52:02 avg_step=1.34s recent_step=1.56s eta=01:18:58 +2026-08-26 15:47:44 | INFO | train rank=0 | step=96470/100000 progress=96.5% loss=0.042587 lr=3.402896e-08 elapsed=35:52:14 avg_step=1.34s recent_step=1.22s eta=01:18:45 +2026-08-26 15:48:00 | INFO | train rank=0 | step=96480/100000 progress=96.5% loss=0.045471 lr=3.383666e-08 elapsed=35:52:30 avg_step=1.34s recent_step=1.62s eta=01:18:31 +2026-08-26 15:48:19 | INFO | train rank=0 | step=96490/100000 progress=96.5% loss=0.044281 lr=3.364489e-08 elapsed=35:52:49 avg_step=1.34s recent_step=1.91s eta=01:18:18 +2026-08-26 15:48:31 | INFO | train rank=0 | step=96500/100000 progress=96.5% loss=0.054123 lr=3.345367e-08 elapsed=35:53:01 avg_step=1.34s recent_step=1.19s eta=01:18:05 +2026-08-26 15:48:46 | INFO | train rank=0 | step=96510/100000 progress=96.5% loss=0.035217 lr=3.326299e-08 elapsed=35:53:15 avg_step=1.34s recent_step=1.44s eta=01:17:51 +2026-08-26 15:49:00 | INFO | train rank=0 | step=96520/100000 progress=96.5% loss=0.049454 lr=3.307285e-08 elapsed=35:53:30 avg_step=1.34s recent_step=1.44s eta=01:17:38 +2026-08-26 15:49:13 | INFO | train rank=0 | step=96530/100000 progress=96.5% loss=0.044121 lr=3.288326e-08 elapsed=35:53:43 avg_step=1.34s recent_step=1.33s eta=01:17:25 +2026-08-26 15:49:25 | INFO | train rank=0 | step=96540/100000 progress=96.5% loss=0.039154 lr=3.269421e-08 elapsed=35:53:55 avg_step=1.34s recent_step=1.19s eta=01:17:11 +2026-08-26 15:49:37 | INFO | train rank=0 | step=96550/100000 progress=96.5% loss=0.063965 lr=3.250571e-08 elapsed=35:54:07 avg_step=1.34s recent_step=1.22s eta=01:16:58 +2026-08-26 15:49:49 | INFO | train rank=0 | step=96560/100000 progress=96.6% loss=0.052475 lr=3.231774e-08 elapsed=35:54:19 avg_step=1.34s recent_step=1.12s eta=01:16:44 +2026-08-26 15:50:06 | INFO | train rank=0 | step=96570/100000 progress=96.6% loss=0.036102 lr=3.213032e-08 elapsed=35:54:36 avg_step=1.34s recent_step=1.74s eta=01:16:31 +2026-08-26 15:50:18 | INFO | train rank=0 | step=96580/100000 progress=96.6% loss=0.052124 lr=3.194345e-08 elapsed=35:54:48 avg_step=1.34s recent_step=1.16s eta=01:16:18 +2026-08-26 15:50:32 | INFO | train rank=0 | step=96590/100000 progress=96.6% loss=0.052948 lr=3.175711e-08 elapsed=35:55:02 avg_step=1.34s recent_step=1.47s eta=01:16:04 +2026-08-26 15:50:48 | INFO | train rank=0 | step=96600/100000 progress=96.6% loss=0.040146 lr=3.157132e-08 elapsed=35:55:18 avg_step=1.34s recent_step=1.55s eta=01:15:51 +2026-08-26 15:51:01 | INFO | train rank=0 | step=96610/100000 progress=96.6% loss=0.048836 lr=3.138608e-08 elapsed=35:55:31 avg_step=1.34s recent_step=1.30s eta=01:15:38 +2026-08-26 15:51:12 | INFO | train rank=0 | step=96620/100000 progress=96.6% loss=0.035400 lr=3.120138e-08 elapsed=35:55:42 avg_step=1.34s recent_step=1.14s eta=01:15:24 +2026-08-26 15:51:25 | INFO | train rank=0 | step=96630/100000 progress=96.6% loss=0.058136 lr=3.101722e-08 elapsed=35:55:55 avg_step=1.34s recent_step=1.25s eta=01:15:11 +2026-08-26 15:51:39 | INFO | train rank=0 | step=96640/100000 progress=96.6% loss=0.036171 lr=3.083360e-08 elapsed=35:56:09 avg_step=1.34s recent_step=1.41s eta=01:14:57 +2026-08-26 15:51:53 | INFO | train rank=0 | step=96650/100000 progress=96.7% loss=0.058395 lr=3.065053e-08 elapsed=35:56:23 avg_step=1.34s recent_step=1.38s eta=01:14:44 +2026-08-26 15:52:04 | INFO | train rank=0 | step=96660/100000 progress=96.7% loss=0.028709 lr=3.046800e-08 elapsed=35:56:34 avg_step=1.34s recent_step=1.17s eta=01:14:31 +2026-08-26 15:52:16 | INFO | train rank=0 | step=96670/100000 progress=96.7% loss=0.036331 lr=3.028601e-08 elapsed=35:56:46 avg_step=1.34s recent_step=1.16s eta=01:14:17 +2026-08-26 15:52:32 | INFO | train rank=0 | step=96680/100000 progress=96.7% loss=0.141479 lr=3.010457e-08 elapsed=35:57:02 avg_step=1.34s recent_step=1.62s eta=01:14:04 +2026-08-26 15:52:43 | INFO | train rank=0 | step=96690/100000 progress=96.7% loss=0.043091 lr=2.992367e-08 elapsed=35:57:13 avg_step=1.34s recent_step=1.11s eta=01:13:50 +2026-08-26 15:52:58 | INFO | train rank=0 | step=96700/100000 progress=96.7% loss=0.045593 lr=2.974332e-08 elapsed=35:57:27 avg_step=1.34s recent_step=1.42s eta=01:13:37 +2026-08-26 15:53:10 | INFO | train rank=0 | step=96710/100000 progress=96.7% loss=0.033981 lr=2.956350e-08 elapsed=35:57:40 avg_step=1.34s recent_step=1.25s eta=01:13:24 +2026-08-26 15:53:21 | INFO | train rank=0 | step=96720/100000 progress=96.7% loss=0.063660 lr=2.938424e-08 elapsed=35:57:51 avg_step=1.34s recent_step=1.12s eta=01:13:10 +2026-08-26 15:53:40 | INFO | train rank=0 | step=96730/100000 progress=96.7% loss=0.060928 lr=2.920551e-08 elapsed=35:58:10 avg_step=1.34s recent_step=1.87s eta=01:12:57 +2026-08-26 15:53:51 | INFO | train rank=0 | step=96740/100000 progress=96.7% loss=0.027580 lr=2.902733e-08 elapsed=35:58:21 avg_step=1.34s recent_step=1.11s eta=01:12:44 +2026-08-26 15:54:06 | INFO | train rank=0 | step=96750/100000 progress=96.8% loss=0.035004 lr=2.884969e-08 elapsed=35:58:36 avg_step=1.34s recent_step=1.51s eta=01:12:30 +2026-08-26 15:54:18 | INFO | train rank=0 | step=96760/100000 progress=96.8% loss=0.057144 lr=2.867260e-08 elapsed=35:58:48 avg_step=1.34s recent_step=1.19s eta=01:12:17 +2026-08-26 15:54:32 | INFO | train rank=0 | step=96770/100000 progress=96.8% loss=0.062195 lr=2.849605e-08 elapsed=35:59:02 avg_step=1.34s recent_step=1.44s eta=01:12:03 +2026-08-26 15:54:48 | INFO | train rank=0 | step=96780/100000 progress=96.8% loss=0.053345 lr=2.832004e-08 elapsed=35:59:18 avg_step=1.34s recent_step=1.58s eta=01:11:50 +2026-08-26 15:55:05 | INFO | train rank=0 | step=96790/100000 progress=96.8% loss=0.043030 lr=2.814458e-08 elapsed=35:59:35 avg_step=1.34s recent_step=1.68s eta=01:11:37 +2026-08-26 15:55:16 | INFO | train rank=0 | step=96800/100000 progress=96.8% loss=0.041061 lr=2.796966e-08 elapsed=35:59:46 avg_step=1.34s recent_step=1.14s eta=01:11:23 +2026-08-26 15:55:30 | INFO | train rank=0 | step=96810/100000 progress=96.8% loss=0.058517 lr=2.779528e-08 elapsed=35:59:59 avg_step=1.34s recent_step=1.31s eta=01:11:10 +2026-08-26 15:55:41 | INFO | train rank=0 | step=96820/100000 progress=96.8% loss=0.044312 lr=2.762145e-08 elapsed=36:00:11 avg_step=1.34s recent_step=1.13s eta=01:10:57 +2026-08-26 15:55:54 | INFO | train rank=0 | step=96830/100000 progress=96.8% loss=0.063660 lr=2.744816e-08 elapsed=36:00:24 avg_step=1.34s recent_step=1.32s eta=01:10:43 +2026-08-26 15:56:09 | INFO | train rank=0 | step=96840/100000 progress=96.8% loss=0.053894 lr=2.727542e-08 elapsed=36:00:39 avg_step=1.34s recent_step=1.54s eta=01:10:30 +2026-08-26 15:56:23 | INFO | train rank=0 | step=96850/100000 progress=96.9% loss=0.047043 lr=2.710322e-08 elapsed=36:00:52 avg_step=1.34s recent_step=1.32s eta=01:10:16 +2026-08-26 15:56:37 | INFO | train rank=0 | step=96860/100000 progress=96.9% loss=0.051338 lr=2.693156e-08 elapsed=36:01:07 avg_step=1.34s recent_step=1.47s eta=01:10:03 +2026-08-26 15:56:49 | INFO | train rank=0 | step=96870/100000 progress=96.9% loss=0.053284 lr=2.676045e-08 elapsed=36:01:19 avg_step=1.34s recent_step=1.18s eta=01:09:50 +2026-08-26 15:57:04 | INFO | train rank=0 | step=96880/100000 progress=96.9% loss=0.036575 lr=2.658988e-08 elapsed=36:01:34 avg_step=1.34s recent_step=1.54s eta=01:09:36 +2026-08-26 15:57:18 | INFO | train rank=0 | step=96890/100000 progress=96.9% loss=0.021389 lr=2.641986e-08 elapsed=36:01:48 avg_step=1.34s recent_step=1.39s eta=01:09:23 +2026-08-26 15:57:30 | INFO | train rank=0 | step=96900/100000 progress=96.9% loss=0.031509 lr=2.625038e-08 elapsed=36:02:00 avg_step=1.34s recent_step=1.17s eta=01:09:09 +2026-08-26 15:57:42 | INFO | train rank=0 | step=96910/100000 progress=96.9% loss=0.095032 lr=2.608144e-08 elapsed=36:02:12 avg_step=1.34s recent_step=1.24s eta=01:08:56 +2026-08-26 15:57:55 | INFO | train rank=0 | step=96920/100000 progress=96.9% loss=0.040909 lr=2.591305e-08 elapsed=36:02:25 avg_step=1.34s recent_step=1.30s eta=01:08:43 +2026-08-26 15:58:08 | INFO | train rank=0 | step=96930/100000 progress=96.9% loss=0.072723 lr=2.574520e-08 elapsed=36:02:37 avg_step=1.34s recent_step=1.21s eta=01:08:29 +2026-08-26 15:58:19 | INFO | train rank=0 | step=96940/100000 progress=96.9% loss=0.062332 lr=2.557789e-08 elapsed=36:02:49 avg_step=1.34s recent_step=1.19s eta=01:08:16 +2026-08-26 15:58:31 | INFO | train rank=0 | step=96950/100000 progress=97.0% loss=0.036613 lr=2.541113e-08 elapsed=36:03:01 avg_step=1.34s recent_step=1.16s eta=01:08:02 +2026-08-26 15:58:45 | INFO | train rank=0 | step=96960/100000 progress=97.0% loss=0.031677 lr=2.524492e-08 elapsed=36:03:14 avg_step=1.34s recent_step=1.35s eta=01:07:49 +2026-08-26 15:58:58 | INFO | train rank=0 | step=96970/100000 progress=97.0% loss=0.048401 lr=2.507924e-08 elapsed=36:03:28 avg_step=1.34s recent_step=1.33s eta=01:07:36 +2026-08-26 15:59:09 | INFO | train rank=0 | step=96980/100000 progress=97.0% loss=0.090149 lr=2.491411e-08 elapsed=36:03:39 avg_step=1.34s recent_step=1.14s eta=01:07:22 +2026-08-26 15:59:21 | INFO | train rank=0 | step=96990/100000 progress=97.0% loss=0.038765 lr=2.474953e-08 elapsed=36:03:50 avg_step=1.34s recent_step=1.13s eta=01:07:09 +2026-08-26 15:59:32 | INFO | train rank=0 | step=97000/100000 progress=97.0% loss=0.049789 lr=2.458549e-08 elapsed=36:04:02 avg_step=1.34s recent_step=1.12s eta=01:06:55 +2026-08-26 15:59:51 | INFO | train rank=0 | step=97010/100000 progress=97.0% loss=0.036766 lr=2.442199e-08 elapsed=36:04:21 avg_step=1.34s recent_step=1.93s eta=01:06:42 +2026-08-26 16:00:03 | INFO | train rank=0 | step=97020/100000 progress=97.0% loss=0.031059 lr=2.425904e-08 elapsed=36:04:33 avg_step=1.34s recent_step=1.21s eta=01:06:29 +2026-08-26 16:00:16 | INFO | train rank=0 | step=97030/100000 progress=97.0% loss=0.043106 lr=2.409663e-08 elapsed=36:04:46 avg_step=1.34s recent_step=1.29s eta=01:06:15 +2026-08-26 16:00:29 | INFO | train rank=0 | step=97040/100000 progress=97.0% loss=0.036491 lr=2.393477e-08 elapsed=36:04:59 avg_step=1.34s recent_step=1.29s eta=01:06:02 +2026-08-26 16:00:44 | INFO | train rank=0 | step=97050/100000 progress=97.0% loss=0.081940 lr=2.377344e-08 elapsed=36:05:14 avg_step=1.34s recent_step=1.49s eta=01:05:48 +2026-08-26 16:00:56 | INFO | train rank=0 | step=97060/100000 progress=97.1% loss=0.060150 lr=2.361267e-08 elapsed=36:05:26 avg_step=1.34s recent_step=1.19s eta=01:05:35 +2026-08-26 16:01:08 | INFO | train rank=0 | step=97070/100000 progress=97.1% loss=0.048599 lr=2.345244e-08 elapsed=36:05:38 avg_step=1.34s recent_step=1.20s eta=01:05:22 +2026-08-26 16:01:23 | INFO | train rank=0 | step=97080/100000 progress=97.1% loss=0.045349 lr=2.329275e-08 elapsed=36:05:53 avg_step=1.34s recent_step=1.50s eta=01:05:08 +2026-08-26 16:01:35 | INFO | train rank=0 | step=97090/100000 progress=97.1% loss=0.028976 lr=2.313361e-08 elapsed=36:06:05 avg_step=1.34s recent_step=1.24s eta=01:04:55 +2026-08-26 16:01:49 | INFO | train rank=0 | step=97100/100000 progress=97.1% loss=0.047501 lr=2.297501e-08 elapsed=36:06:19 avg_step=1.34s recent_step=1.39s eta=01:04:41 +2026-08-26 16:02:03 | INFO | train rank=0 | step=97110/100000 progress=97.1% loss=0.051514 lr=2.281695e-08 elapsed=36:06:33 avg_step=1.34s recent_step=1.43s eta=01:04:28 +2026-08-26 16:02:15 | INFO | train rank=0 | step=97120/100000 progress=97.1% loss=0.043747 lr=2.265944e-08 elapsed=36:06:45 avg_step=1.34s recent_step=1.17s eta=01:04:15 +2026-08-26 16:02:29 | INFO | train rank=0 | step=97130/100000 progress=97.1% loss=0.083954 lr=2.250248e-08 elapsed=36:06:59 avg_step=1.34s recent_step=1.40s eta=01:04:01 +2026-08-26 16:02:41 | INFO | train rank=0 | step=97140/100000 progress=97.1% loss=0.069214 lr=2.234605e-08 elapsed=36:07:11 avg_step=1.34s recent_step=1.16s eta=01:03:48 +2026-08-26 16:02:55 | INFO | train rank=0 | step=97150/100000 progress=97.2% loss=0.041229 lr=2.219018e-08 elapsed=36:07:25 avg_step=1.34s recent_step=1.43s eta=01:03:35 +2026-08-26 16:03:07 | INFO | train rank=0 | step=97160/100000 progress=97.2% loss=0.050171 lr=2.203484e-08 elapsed=36:07:37 avg_step=1.34s recent_step=1.17s eta=01:03:21 +2026-08-26 16:03:22 | INFO | train rank=0 | step=97170/100000 progress=97.2% loss=0.098755 lr=2.188006e-08 elapsed=36:07:52 avg_step=1.34s recent_step=1.51s eta=01:03:08 +2026-08-26 16:03:37 | INFO | train rank=0 | step=97180/100000 progress=97.2% loss=0.052185 lr=2.172581e-08 elapsed=36:08:06 avg_step=1.34s recent_step=1.47s eta=01:02:54 +2026-08-26 16:03:50 | INFO | train rank=0 | step=97190/100000 progress=97.2% loss=0.034386 lr=2.157211e-08 elapsed=36:08:20 avg_step=1.34s recent_step=1.37s eta=01:02:41 +2026-08-26 16:04:04 | INFO | train rank=0 | step=97200/100000 progress=97.2% loss=0.105530 lr=2.141896e-08 elapsed=36:08:34 avg_step=1.34s recent_step=1.38s eta=01:02:28 +2026-08-26 16:04:18 | INFO | train rank=0 | step=97210/100000 progress=97.2% loss=0.043442 lr=2.126634e-08 elapsed=36:08:48 avg_step=1.34s recent_step=1.42s eta=01:02:14 +2026-08-26 16:04:32 | INFO | train rank=0 | step=97220/100000 progress=97.2% loss=0.037918 lr=2.111428e-08 elapsed=36:09:02 avg_step=1.34s recent_step=1.41s eta=01:02:01 +2026-08-26 16:04:48 | INFO | train rank=0 | step=97230/100000 progress=97.2% loss=0.044510 lr=2.096276e-08 elapsed=36:09:18 avg_step=1.34s recent_step=1.60s eta=01:01:48 +2026-08-26 16:05:00 | INFO | train rank=0 | step=97240/100000 progress=97.2% loss=0.033829 lr=2.081178e-08 elapsed=36:09:30 avg_step=1.34s recent_step=1.12s eta=01:01:34 +2026-08-26 16:05:11 | INFO | train rank=0 | step=97250/100000 progress=97.2% loss=0.063934 lr=2.066135e-08 elapsed=36:09:41 avg_step=1.34s recent_step=1.15s eta=01:01:21 +2026-08-26 16:05:25 | INFO | train rank=0 | step=97260/100000 progress=97.3% loss=0.072418 lr=2.051146e-08 elapsed=36:09:54 avg_step=1.34s recent_step=1.34s eta=01:01:07 +2026-08-26 16:05:42 | INFO | train rank=0 | step=97270/100000 progress=97.3% loss=0.045242 lr=2.036211e-08 elapsed=36:10:11 avg_step=1.34s recent_step=1.70s eta=01:00:54 +2026-08-26 16:05:54 | INFO | train rank=0 | step=97280/100000 progress=97.3% loss=0.054184 lr=2.021331e-08 elapsed=36:10:24 avg_step=1.34s recent_step=1.26s eta=01:00:41 +2026-08-26 16:06:06 | INFO | train rank=0 | step=97290/100000 progress=97.3% loss=0.042496 lr=2.006506e-08 elapsed=36:10:36 avg_step=1.34s recent_step=1.20s eta=01:00:27 +2026-08-26 16:06:22 | INFO | train rank=0 | step=97300/100000 progress=97.3% loss=0.039825 lr=1.991735e-08 elapsed=36:10:52 avg_step=1.34s recent_step=1.57s eta=01:00:14 +2026-08-26 16:06:35 | INFO | train rank=0 | step=97310/100000 progress=97.3% loss=0.050056 lr=1.977018e-08 elapsed=36:11:05 avg_step=1.34s recent_step=1.31s eta=01:00:01 +2026-08-26 16:06:48 | INFO | train rank=0 | step=97320/100000 progress=97.3% loss=0.075836 lr=1.962356e-08 elapsed=36:11:18 avg_step=1.34s recent_step=1.34s eta=00:59:47 +2026-08-26 16:07:00 | INFO | train rank=0 | step=97330/100000 progress=97.3% loss=0.065765 lr=1.947749e-08 elapsed=36:11:30 avg_step=1.34s recent_step=1.18s eta=00:59:34 +2026-08-26 16:07:18 | INFO | train rank=0 | step=97340/100000 progress=97.3% loss=0.040787 lr=1.933195e-08 elapsed=36:11:48 avg_step=1.34s recent_step=1.76s eta=00:59:20 +2026-08-26 16:07:32 | INFO | train rank=0 | step=97350/100000 progress=97.4% loss=0.030144 lr=1.918697e-08 elapsed=36:12:02 avg_step=1.34s recent_step=1.39s eta=00:59:07 +2026-08-26 16:07:44 | INFO | train rank=0 | step=97360/100000 progress=97.4% loss=0.040520 lr=1.904253e-08 elapsed=36:12:14 avg_step=1.34s recent_step=1.26s eta=00:58:54 +2026-08-26 16:08:01 | INFO | train rank=0 | step=97370/100000 progress=97.4% loss=0.037750 lr=1.889863e-08 elapsed=36:12:31 avg_step=1.34s recent_step=1.70s eta=00:58:40 +2026-08-26 16:08:13 | INFO | train rank=0 | step=97380/100000 progress=97.4% loss=0.051163 lr=1.875527e-08 elapsed=36:12:42 avg_step=1.34s recent_step=1.13s eta=00:58:27 +2026-08-26 16:08:29 | INFO | train rank=0 | step=97390/100000 progress=97.4% loss=0.070404 lr=1.861247e-08 elapsed=36:12:59 avg_step=1.34s recent_step=1.61s eta=00:58:14 +2026-08-26 16:08:41 | INFO | train rank=0 | step=97400/100000 progress=97.4% loss=0.037445 lr=1.847020e-08 elapsed=36:13:11 avg_step=1.34s recent_step=1.22s eta=00:58:00 +2026-08-26 16:08:56 | INFO | train rank=0 | step=97410/100000 progress=97.4% loss=0.035275 lr=1.832848e-08 elapsed=36:13:26 avg_step=1.34s recent_step=1.49s eta=00:57:47 +2026-08-26 16:09:08 | INFO | train rank=0 | step=97420/100000 progress=97.4% loss=0.038864 lr=1.818731e-08 elapsed=36:13:37 avg_step=1.34s recent_step=1.18s eta=00:57:33 +2026-08-26 16:09:19 | INFO | train rank=0 | step=97430/100000 progress=97.4% loss=0.043381 lr=1.804668e-08 elapsed=36:13:49 avg_step=1.34s recent_step=1.16s eta=00:57:20 +2026-08-26 16:09:31 | INFO | train rank=0 | step=97440/100000 progress=97.4% loss=0.025124 lr=1.790660e-08 elapsed=36:14:01 avg_step=1.34s recent_step=1.22s eta=00:57:07 +2026-08-26 16:09:48 | INFO | train rank=0 | step=97450/100000 progress=97.5% loss=0.041107 lr=1.776706e-08 elapsed=36:14:18 avg_step=1.34s recent_step=1.63s eta=00:56:53 +2026-08-26 16:10:00 | INFO | train rank=0 | step=97460/100000 progress=97.5% loss=0.042603 lr=1.762806e-08 elapsed=36:14:29 avg_step=1.34s recent_step=1.18s eta=00:56:40 +2026-08-26 16:10:15 | INFO | train rank=0 | step=97470/100000 progress=97.5% loss=0.049713 lr=1.748961e-08 elapsed=36:14:45 avg_step=1.34s recent_step=1.57s eta=00:56:26 +2026-08-26 16:10:27 | INFO | train rank=0 | step=97480/100000 progress=97.5% loss=0.031563 lr=1.735171e-08 elapsed=36:14:57 avg_step=1.34s recent_step=1.15s eta=00:56:13 +2026-08-26 16:10:39 | INFO | train rank=0 | step=97490/100000 progress=97.5% loss=0.044342 lr=1.721435e-08 elapsed=36:15:09 avg_step=1.34s recent_step=1.21s eta=00:56:00 +2026-08-26 16:10:54 | INFO | train rank=0 | step=97500/100000 progress=97.5% loss=0.045792 lr=1.707753e-08 elapsed=36:15:24 avg_step=1.34s recent_step=1.50s eta=00:55:46 +2026-08-26 16:11:10 | INFO | train rank=0 | step=97510/100000 progress=97.5% loss=0.035202 lr=1.694126e-08 elapsed=36:15:40 avg_step=1.34s recent_step=1.66s eta=00:55:33 +2026-08-26 16:11:22 | INFO | train rank=0 | step=97520/100000 progress=97.5% loss=0.066925 lr=1.680554e-08 elapsed=36:15:52 avg_step=1.34s recent_step=1.15s eta=00:55:20 +2026-08-26 16:11:35 | INFO | train rank=0 | step=97530/100000 progress=97.5% loss=0.040443 lr=1.667036e-08 elapsed=36:16:05 avg_step=1.34s recent_step=1.27s eta=00:55:06 +2026-08-26 16:11:48 | INFO | train rank=0 | step=97540/100000 progress=97.5% loss=0.041397 lr=1.653572e-08 elapsed=36:16:18 avg_step=1.34s recent_step=1.38s eta=00:54:53 +2026-08-26 16:12:00 | INFO | train rank=0 | step=97550/100000 progress=97.5% loss=0.055809 lr=1.640163e-08 elapsed=36:16:30 avg_step=1.34s recent_step=1.14s eta=00:54:39 +2026-08-26 16:12:12 | INFO | train rank=0 | step=97560/100000 progress=97.6% loss=0.042900 lr=1.626809e-08 elapsed=36:16:42 avg_step=1.34s recent_step=1.19s eta=00:54:26 +2026-08-26 16:12:24 | INFO | train rank=0 | step=97570/100000 progress=97.6% loss=0.047653 lr=1.613509e-08 elapsed=36:16:54 avg_step=1.34s recent_step=1.25s eta=00:54:12 +2026-08-26 16:12:36 | INFO | train rank=0 | step=97580/100000 progress=97.6% loss=0.125793 lr=1.600263e-08 elapsed=36:17:06 avg_step=1.34s recent_step=1.15s eta=00:53:59 +2026-08-26 16:12:54 | INFO | train rank=0 | step=97590/100000 progress=97.6% loss=0.058517 lr=1.587072e-08 elapsed=36:17:24 avg_step=1.34s recent_step=1.81s eta=00:53:46 +2026-08-26 16:13:09 | INFO | train rank=0 | step=97600/100000 progress=97.6% loss=0.044128 lr=1.573936e-08 elapsed=36:17:39 avg_step=1.34s recent_step=1.55s eta=00:53:32 +2026-08-26 16:13:23 | INFO | train rank=0 | step=97610/100000 progress=97.6% loss=0.044731 lr=1.560854e-08 elapsed=36:17:52 avg_step=1.34s recent_step=1.32s eta=00:53:19 +2026-08-26 16:13:36 | INFO | train rank=0 | step=97620/100000 progress=97.6% loss=0.053253 lr=1.547826e-08 elapsed=36:18:06 avg_step=1.34s recent_step=1.33s eta=00:53:06 +2026-08-26 16:13:53 | INFO | train rank=0 | step=97630/100000 progress=97.6% loss=0.040634 lr=1.534853e-08 elapsed=36:18:23 avg_step=1.34s recent_step=1.71s eta=00:52:52 +2026-08-26 16:14:07 | INFO | train rank=0 | step=97640/100000 progress=97.6% loss=0.086182 lr=1.521935e-08 elapsed=36:18:37 avg_step=1.34s recent_step=1.38s eta=00:52:39 +2026-08-26 16:14:19 | INFO | train rank=0 | step=97650/100000 progress=97.7% loss=0.035507 lr=1.509071e-08 elapsed=36:18:49 avg_step=1.34s recent_step=1.26s eta=00:52:26 +2026-08-26 16:14:30 | INFO | train rank=0 | step=97660/100000 progress=97.7% loss=0.048340 lr=1.496262e-08 elapsed=36:19:00 avg_step=1.34s recent_step=1.10s eta=00:52:12 +2026-08-26 16:14:46 | INFO | train rank=0 | step=97670/100000 progress=97.7% loss=0.042404 lr=1.483507e-08 elapsed=36:19:16 avg_step=1.34s recent_step=1.56s eta=00:51:59 +2026-08-26 16:15:04 | INFO | train rank=0 | step=97680/100000 progress=97.7% loss=0.077698 lr=1.470806e-08 elapsed=36:19:34 avg_step=1.34s recent_step=1.83s eta=00:51:46 +2026-08-26 16:15:16 | INFO | train rank=0 | step=97690/100000 progress=97.7% loss=0.038185 lr=1.458160e-08 elapsed=36:19:46 avg_step=1.34s recent_step=1.15s eta=00:51:32 +2026-08-26 16:15:28 | INFO | train rank=0 | step=97700/100000 progress=97.7% loss=0.039032 lr=1.445569e-08 elapsed=36:19:58 avg_step=1.34s recent_step=1.26s eta=00:51:19 +2026-08-26 16:15:47 | INFO | train rank=0 | step=97710/100000 progress=97.7% loss=0.045837 lr=1.433032e-08 elapsed=36:20:16 avg_step=1.34s recent_step=1.82s eta=00:51:05 +2026-08-26 16:15:58 | INFO | train rank=0 | step=97720/100000 progress=97.7% loss=0.039368 lr=1.420550e-08 elapsed=36:20:28 avg_step=1.34s recent_step=1.19s eta=00:50:52 +2026-08-26 16:16:12 | INFO | train rank=0 | step=97730/100000 progress=97.7% loss=0.041153 lr=1.408122e-08 elapsed=36:20:42 avg_step=1.34s recent_step=1.39s eta=00:50:39 +2026-08-26 16:16:24 | INFO | train rank=0 | step=97740/100000 progress=97.7% loss=0.050598 lr=1.395749e-08 elapsed=36:20:54 avg_step=1.34s recent_step=1.17s eta=00:50:25 +2026-08-26 16:16:40 | INFO | train rank=0 | step=97750/100000 progress=97.8% loss=0.034790 lr=1.383430e-08 elapsed=36:21:10 avg_step=1.34s recent_step=1.60s eta=00:50:12 +2026-08-26 16:16:53 | INFO | train rank=0 | step=97760/100000 progress=97.8% loss=0.054153 lr=1.371166e-08 elapsed=36:21:23 avg_step=1.34s recent_step=1.34s eta=00:49:58 +2026-08-26 16:17:07 | INFO | train rank=0 | step=97770/100000 progress=97.8% loss=0.043602 lr=1.358956e-08 elapsed=36:21:37 avg_step=1.34s recent_step=1.34s eta=00:49:45 +2026-08-26 16:17:21 | INFO | train rank=0 | step=97780/100000 progress=97.8% loss=0.067177 lr=1.346801e-08 elapsed=36:21:51 avg_step=1.34s recent_step=1.43s eta=00:49:32 +2026-08-26 16:17:32 | INFO | train rank=0 | step=97790/100000 progress=97.8% loss=0.033028 lr=1.334700e-08 elapsed=36:22:02 avg_step=1.34s recent_step=1.09s eta=00:49:18 +2026-08-26 16:17:45 | INFO | train rank=0 | step=97800/100000 progress=97.8% loss=0.055908 lr=1.322654e-08 elapsed=36:22:15 avg_step=1.34s recent_step=1.34s eta=00:49:05 +2026-08-26 16:17:57 | INFO | train rank=0 | step=97810/100000 progress=97.8% loss=0.044098 lr=1.310663e-08 elapsed=36:22:27 avg_step=1.34s recent_step=1.13s eta=00:48:51 +2026-08-26 16:18:08 | INFO | train rank=0 | step=97820/100000 progress=97.8% loss=0.047180 lr=1.298726e-08 elapsed=36:22:38 avg_step=1.34s recent_step=1.10s eta=00:48:38 +2026-08-26 16:18:27 | INFO | train rank=0 | step=97830/100000 progress=97.8% loss=0.043793 lr=1.286843e-08 elapsed=36:22:57 avg_step=1.34s recent_step=1.98s eta=00:48:25 +2026-08-26 16:18:39 | INFO | train rank=0 | step=97840/100000 progress=97.8% loss=0.064926 lr=1.275015e-08 elapsed=36:23:09 avg_step=1.34s recent_step=1.12s eta=00:48:11 +2026-08-26 16:18:53 | INFO | train rank=0 | step=97850/100000 progress=97.9% loss=0.057785 lr=1.263242e-08 elapsed=36:23:23 avg_step=1.34s recent_step=1.47s eta=00:47:58 +2026-08-26 16:19:05 | INFO | train rank=0 | step=97860/100000 progress=97.9% loss=0.050491 lr=1.251523e-08 elapsed=36:23:35 avg_step=1.34s recent_step=1.21s eta=00:47:45 +2026-08-26 16:19:17 | INFO | train rank=0 | step=97870/100000 progress=97.9% loss=0.049683 lr=1.239859e-08 elapsed=36:23:47 avg_step=1.34s recent_step=1.20s eta=00:47:31 +2026-08-26 16:19:30 | INFO | train rank=0 | step=97880/100000 progress=97.9% loss=0.051758 lr=1.228249e-08 elapsed=36:23:59 avg_step=1.34s recent_step=1.20s eta=00:47:18 +2026-08-26 16:19:42 | INFO | train rank=0 | step=97890/100000 progress=97.9% loss=0.094025 lr=1.216694e-08 elapsed=36:24:12 avg_step=1.34s recent_step=1.23s eta=00:47:04 +2026-08-26 16:19:53 | INFO | train rank=0 | step=97900/100000 progress=97.9% loss=0.059921 lr=1.205193e-08 elapsed=36:24:23 avg_step=1.34s recent_step=1.10s eta=00:46:51 +2026-08-26 16:20:07 | INFO | train rank=0 | step=97910/100000 progress=97.9% loss=0.038101 lr=1.193747e-08 elapsed=36:24:37 avg_step=1.34s recent_step=1.39s eta=00:46:37 +2026-08-26 16:20:18 | INFO | train rank=0 | step=97920/100000 progress=97.9% loss=0.027763 lr=1.182355e-08 elapsed=36:24:48 avg_step=1.34s recent_step=1.11s eta=00:46:24 +2026-08-26 16:20:33 | INFO | train rank=0 | step=97930/100000 progress=97.9% loss=0.040783 lr=1.171018e-08 elapsed=36:25:03 avg_step=1.34s recent_step=1.50s eta=00:46:11 +2026-08-26 16:20:49 | INFO | train rank=0 | step=97940/100000 progress=97.9% loss=0.049164 lr=1.159736e-08 elapsed=36:25:19 avg_step=1.34s recent_step=1.63s eta=00:45:57 +2026-08-26 16:21:03 | INFO | train rank=0 | step=97950/100000 progress=98.0% loss=0.060379 lr=1.148508e-08 elapsed=36:25:33 avg_step=1.34s recent_step=1.40s eta=00:45:44 +2026-08-26 16:21:18 | INFO | train rank=0 | step=97960/100000 progress=98.0% loss=0.059326 lr=1.137334e-08 elapsed=36:25:48 avg_step=1.34s recent_step=1.47s eta=00:45:31 +2026-08-26 16:21:29 | INFO | train rank=0 | step=97970/100000 progress=98.0% loss=0.059387 lr=1.126216e-08 elapsed=36:25:59 avg_step=1.34s recent_step=1.12s eta=00:45:17 +2026-08-26 16:21:41 | INFO | train rank=0 | step=97980/100000 progress=98.0% loss=0.054184 lr=1.115151e-08 elapsed=36:26:11 avg_step=1.34s recent_step=1.18s eta=00:45:04 +2026-08-26 16:21:53 | INFO | train rank=0 | step=97990/100000 progress=98.0% loss=0.042603 lr=1.104142e-08 elapsed=36:26:23 avg_step=1.34s recent_step=1.25s eta=00:44:50 +2026-08-26 16:22:09 | INFO | train rank=0 | step=98000/100000 progress=98.0% loss=0.059402 lr=1.093186e-08 elapsed=36:26:39 avg_step=1.34s recent_step=1.57s eta=00:44:37 +2026-08-26 16:22:21 | INFO | train rank=0 | step=98010/100000 progress=98.0% loss=0.042137 lr=1.082286e-08 elapsed=36:26:51 avg_step=1.34s recent_step=1.16s eta=00:44:24 +2026-08-26 16:22:37 | INFO | train rank=0 | step=98020/100000 progress=98.0% loss=0.027504 lr=1.071440e-08 elapsed=36:27:07 avg_step=1.34s recent_step=1.61s eta=00:44:10 +2026-08-26 16:22:51 | INFO | train rank=0 | step=98030/100000 progress=98.0% loss=0.042252 lr=1.060648e-08 elapsed=36:27:20 avg_step=1.34s recent_step=1.37s eta=00:43:57 +2026-08-26 16:23:03 | INFO | train rank=0 | step=98040/100000 progress=98.0% loss=0.052704 lr=1.049911e-08 elapsed=36:27:33 avg_step=1.34s recent_step=1.21s eta=00:43:43 +2026-08-26 16:23:17 | INFO | train rank=0 | step=98050/100000 progress=98.0% loss=0.033813 lr=1.039229e-08 elapsed=36:27:47 avg_step=1.34s recent_step=1.44s eta=00:43:30 +2026-08-26 16:23:32 | INFO | train rank=0 | step=98060/100000 progress=98.1% loss=0.033630 lr=1.028601e-08 elapsed=36:28:02 avg_step=1.34s recent_step=1.50s eta=00:43:17 +2026-08-26 16:23:45 | INFO | train rank=0 | step=98070/100000 progress=98.1% loss=0.036797 lr=1.018028e-08 elapsed=36:28:15 avg_step=1.34s recent_step=1.30s eta=00:43:03 +2026-08-26 16:23:58 | INFO | train rank=0 | step=98080/100000 progress=98.1% loss=0.048584 lr=1.007509e-08 elapsed=36:28:28 avg_step=1.34s recent_step=1.27s eta=00:42:50 +2026-08-26 16:24:09 | INFO | train rank=0 | step=98090/100000 progress=98.1% loss=0.035240 lr=9.970453e-09 elapsed=36:28:39 avg_step=1.34s recent_step=1.17s eta=00:42:37 +2026-08-26 16:24:24 | INFO | train rank=0 | step=98100/100000 progress=98.1% loss=0.031509 lr=9.866358e-09 elapsed=36:28:54 avg_step=1.34s recent_step=1.48s eta=00:42:23 +2026-08-26 16:24:41 | INFO | train rank=0 | step=98110/100000 progress=98.1% loss=0.048363 lr=9.762808e-09 elapsed=36:29:11 avg_step=1.34s recent_step=1.64s eta=00:42:10 +2026-08-26 16:24:56 | INFO | train rank=0 | step=98120/100000 progress=98.1% loss=0.039040 lr=9.659805e-09 elapsed=36:29:26 avg_step=1.34s recent_step=1.53s eta=00:41:57 +2026-08-26 16:25:07 | INFO | train rank=0 | step=98130/100000 progress=98.1% loss=0.044762 lr=9.557347e-09 elapsed=36:29:37 avg_step=1.34s recent_step=1.13s eta=00:41:43 +2026-08-26 16:25:21 | INFO | train rank=0 | step=98140/100000 progress=98.1% loss=0.054306 lr=9.455435e-09 elapsed=36:29:50 avg_step=1.34s recent_step=1.34s eta=00:41:30 +2026-08-26 16:25:36 | INFO | train rank=0 | step=98150/100000 progress=98.2% loss=0.024445 lr=9.354068e-09 elapsed=36:30:06 avg_step=1.34s recent_step=1.55s eta=00:41:16 +2026-08-26 16:25:50 | INFO | train rank=0 | step=98160/100000 progress=98.2% loss=0.054169 lr=9.253248e-09 elapsed=36:30:19 avg_step=1.34s recent_step=1.35s eta=00:41:03 +2026-08-26 16:26:01 | INFO | train rank=0 | step=98170/100000 progress=98.2% loss=0.042801 lr=9.152973e-09 elapsed=36:30:31 avg_step=1.34s recent_step=1.17s eta=00:40:50 +2026-08-26 16:26:17 | INFO | train rank=0 | step=98180/100000 progress=98.2% loss=0.028755 lr=9.053244e-09 elapsed=36:30:46 avg_step=1.34s recent_step=1.52s eta=00:40:36 +2026-08-26 16:26:28 | INFO | train rank=0 | step=98190/100000 progress=98.2% loss=0.046707 lr=8.954061e-09 elapsed=36:30:58 avg_step=1.34s recent_step=1.14s eta=00:40:23 +2026-08-26 16:26:40 | INFO | train rank=0 | step=98200/100000 progress=98.2% loss=0.034363 lr=8.855423e-09 elapsed=36:31:10 avg_step=1.34s recent_step=1.17s eta=00:40:09 +2026-08-26 16:26:53 | INFO | train rank=0 | step=98210/100000 progress=98.2% loss=0.055176 lr=8.757331e-09 elapsed=36:31:23 avg_step=1.34s recent_step=1.36s eta=00:39:56 +2026-08-26 16:27:06 | INFO | train rank=0 | step=98220/100000 progress=98.2% loss=0.034760 lr=8.659786e-09 elapsed=36:31:35 avg_step=1.34s recent_step=1.22s eta=00:39:43 +2026-08-26 16:27:27 | INFO | train rank=0 | step=98230/100000 progress=98.2% loss=0.050568 lr=8.562786e-09 elapsed=36:31:56 avg_step=1.34s recent_step=2.10s eta=00:39:29 +2026-08-26 16:27:38 | INFO | train rank=0 | step=98240/100000 progress=98.2% loss=0.041565 lr=8.466332e-09 elapsed=36:32:08 avg_step=1.34s recent_step=1.14s eta=00:39:16 +2026-08-26 16:27:53 | INFO | train rank=0 | step=98250/100000 progress=98.2% loss=0.044327 lr=8.370423e-09 elapsed=36:32:22 avg_step=1.34s recent_step=1.46s eta=00:39:03 +2026-08-26 16:28:05 | INFO | train rank=0 | step=98260/100000 progress=98.3% loss=0.074799 lr=8.275061e-09 elapsed=36:32:34 avg_step=1.34s recent_step=1.20s eta=00:38:49 +2026-08-26 16:28:20 | INFO | train rank=0 | step=98270/100000 progress=98.3% loss=0.044205 lr=8.180245e-09 elapsed=36:32:50 avg_step=1.34s recent_step=1.56s eta=00:38:36 +2026-08-26 16:28:32 | INFO | train rank=0 | step=98280/100000 progress=98.3% loss=0.057388 lr=8.085974e-09 elapsed=36:33:02 avg_step=1.34s recent_step=1.18s eta=00:38:22 +2026-08-26 16:28:46 | INFO | train rank=0 | step=98290/100000 progress=98.3% loss=0.047760 lr=7.992249e-09 elapsed=36:33:16 avg_step=1.34s recent_step=1.41s eta=00:38:09 +2026-08-26 16:29:01 | INFO | train rank=0 | step=98300/100000 progress=98.3% loss=0.040070 lr=7.899071e-09 elapsed=36:33:31 avg_step=1.34s recent_step=1.54s eta=00:37:56 +2026-08-26 16:29:13 | INFO | train rank=0 | step=98310/100000 progress=98.3% loss=0.048141 lr=7.806438e-09 elapsed=36:33:43 avg_step=1.34s recent_step=1.17s eta=00:37:42 +2026-08-26 16:29:25 | INFO | train rank=0 | step=98320/100000 progress=98.3% loss=0.057739 lr=7.714351e-09 elapsed=36:33:55 avg_step=1.34s recent_step=1.17s eta=00:37:29 +2026-08-26 16:29:39 | INFO | train rank=0 | step=98330/100000 progress=98.3% loss=0.035767 lr=7.622810e-09 elapsed=36:34:09 avg_step=1.34s recent_step=1.42s eta=00:37:15 +2026-08-26 16:29:54 | INFO | train rank=0 | step=98340/100000 progress=98.3% loss=0.031197 lr=7.531815e-09 elapsed=36:34:23 avg_step=1.34s recent_step=1.45s eta=00:37:02 +2026-08-26 16:30:10 | INFO | train rank=0 | step=98350/100000 progress=98.4% loss=0.031799 lr=7.441366e-09 elapsed=36:34:40 avg_step=1.34s recent_step=1.68s eta=00:36:49 +2026-08-26 16:30:22 | INFO | train rank=0 | step=98360/100000 progress=98.4% loss=0.047272 lr=7.351463e-09 elapsed=36:34:52 avg_step=1.34s recent_step=1.16s eta=00:36:35 +2026-08-26 16:30:34 | INFO | train rank=0 | step=98370/100000 progress=98.4% loss=0.067459 lr=7.262106e-09 elapsed=36:35:04 avg_step=1.34s recent_step=1.24s eta=00:36:22 +2026-08-26 16:30:46 | INFO | train rank=0 | step=98380/100000 progress=98.4% loss=0.044815 lr=7.173295e-09 elapsed=36:35:16 avg_step=1.34s recent_step=1.19s eta=00:36:08 +2026-08-26 16:30:58 | INFO | train rank=0 | step=98390/100000 progress=98.4% loss=0.053055 lr=7.085030e-09 elapsed=36:35:28 avg_step=1.34s recent_step=1.17s eta=00:35:55 +2026-08-26 16:31:12 | INFO | train rank=0 | step=98400/100000 progress=98.4% loss=0.069885 lr=6.997311e-09 elapsed=36:35:42 avg_step=1.34s recent_step=1.39s eta=00:35:42 +2026-08-26 16:31:24 | INFO | train rank=0 | step=98410/100000 progress=98.4% loss=0.041382 lr=6.910138e-09 elapsed=36:35:54 avg_step=1.34s recent_step=1.19s eta=00:35:28 +2026-08-26 16:31:37 | INFO | train rank=0 | step=98420/100000 progress=98.4% loss=0.046265 lr=6.823511e-09 elapsed=36:36:07 avg_step=1.34s recent_step=1.35s eta=00:35:15 +2026-08-26 16:31:52 | INFO | train rank=0 | step=98430/100000 progress=98.4% loss=0.066956 lr=6.737430e-09 elapsed=36:36:22 avg_step=1.34s recent_step=1.45s eta=00:35:01 +2026-08-26 16:32:07 | INFO | train rank=0 | step=98440/100000 progress=98.4% loss=0.043655 lr=6.651896e-09 elapsed=36:36:37 avg_step=1.34s recent_step=1.50s eta=00:34:48 +2026-08-26 16:32:26 | INFO | train rank=0 | step=98450/100000 progress=98.5% loss=0.033447 lr=6.566907e-09 elapsed=36:36:56 avg_step=1.34s recent_step=1.92s eta=00:34:35 +2026-08-26 16:32:38 | INFO | train rank=0 | step=98460/100000 progress=98.5% loss=0.042969 lr=6.482464e-09 elapsed=36:37:07 avg_step=1.34s recent_step=1.16s eta=00:34:21 +2026-08-26 16:32:49 | INFO | train rank=0 | step=98470/100000 progress=98.5% loss=0.032578 lr=6.398567e-09 elapsed=36:37:19 avg_step=1.34s recent_step=1.18s eta=00:34:08 +2026-08-26 16:33:01 | INFO | train rank=0 | step=98480/100000 progress=98.5% loss=0.036003 lr=6.315217e-09 elapsed=36:37:31 avg_step=1.34s recent_step=1.20s eta=00:33:55 +2026-08-26 16:33:13 | INFO | train rank=0 | step=98490/100000 progress=98.5% loss=0.034500 lr=6.232413e-09 elapsed=36:37:43 avg_step=1.34s recent_step=1.14s eta=00:33:41 +2026-08-26 16:33:26 | INFO | train rank=0 | step=98500/100000 progress=98.5% loss=0.069092 lr=6.150154e-09 elapsed=36:37:56 avg_step=1.34s recent_step=1.36s eta=00:33:28 +2026-08-26 16:33:40 | INFO | train rank=0 | step=98510/100000 progress=98.5% loss=0.065765 lr=6.068442e-09 elapsed=36:38:09 avg_step=1.34s recent_step=1.32s eta=00:33:14 +2026-08-26 16:33:54 | INFO | train rank=0 | step=98520/100000 progress=98.5% loss=0.062256 lr=5.987276e-09 elapsed=36:38:24 avg_step=1.34s recent_step=1.45s eta=00:33:01 +2026-08-26 16:34:06 | INFO | train rank=0 | step=98530/100000 progress=98.5% loss=0.036713 lr=5.906656e-09 elapsed=36:38:36 avg_step=1.34s recent_step=1.25s eta=00:32:48 +2026-08-26 16:34:22 | INFO | train rank=0 | step=98540/100000 progress=98.5% loss=0.073029 lr=5.826582e-09 elapsed=36:38:52 avg_step=1.34s recent_step=1.52s eta=00:32:34 +2026-08-26 16:34:33 | INFO | train rank=0 | step=98550/100000 progress=98.6% loss=0.041527 lr=5.747055e-09 elapsed=36:39:03 avg_step=1.34s recent_step=1.17s eta=00:32:21 +2026-08-26 16:34:47 | INFO | train rank=0 | step=98560/100000 progress=98.6% loss=0.051208 lr=5.668073e-09 elapsed=36:39:17 avg_step=1.34s recent_step=1.37s eta=00:32:07 +2026-08-26 16:35:03 | INFO | train rank=0 | step=98570/100000 progress=98.6% loss=0.037796 lr=5.589638e-09 elapsed=36:39:33 avg_step=1.34s recent_step=1.64s eta=00:31:54 +2026-08-26 16:35:15 | INFO | train rank=0 | step=98580/100000 progress=98.6% loss=0.059647 lr=5.511749e-09 elapsed=36:39:44 avg_step=1.34s recent_step=1.11s eta=00:31:41 +2026-08-26 16:35:26 | INFO | train rank=0 | step=98590/100000 progress=98.6% loss=0.030861 lr=5.434406e-09 elapsed=36:39:56 avg_step=1.34s recent_step=1.17s eta=00:31:27 +2026-08-26 16:35:41 | INFO | train rank=0 | step=98600/100000 progress=98.6% loss=0.047318 lr=5.357609e-09 elapsed=36:40:11 avg_step=1.34s recent_step=1.52s eta=00:31:14 +2026-08-26 16:35:57 | INFO | train rank=0 | step=98610/100000 progress=98.6% loss=0.046600 lr=5.281359e-09 elapsed=36:40:27 avg_step=1.34s recent_step=1.53s eta=00:31:01 +2026-08-26 16:36:08 | INFO | train rank=0 | step=98620/100000 progress=98.6% loss=0.027950 lr=5.205655e-09 elapsed=36:40:38 avg_step=1.34s recent_step=1.14s eta=00:30:47 +2026-08-26 16:36:20 | INFO | train rank=0 | step=98630/100000 progress=98.6% loss=0.035446 lr=5.130496e-09 elapsed=36:40:50 avg_step=1.34s recent_step=1.17s eta=00:30:34 +2026-08-26 16:36:33 | INFO | train rank=0 | step=98640/100000 progress=98.6% loss=0.040939 lr=5.055885e-09 elapsed=36:41:03 avg_step=1.34s recent_step=1.32s eta=00:30:20 +2026-08-26 16:36:49 | INFO | train rank=0 | step=98650/100000 progress=98.7% loss=0.065826 lr=4.981819e-09 elapsed=36:41:19 avg_step=1.34s recent_step=1.62s eta=00:30:07 +2026-08-26 16:37:03 | INFO | train rank=0 | step=98660/100000 progress=98.7% loss=0.032066 lr=4.908300e-09 elapsed=36:41:33 avg_step=1.34s recent_step=1.34s eta=00:29:54 +2026-08-26 16:37:14 | INFO | train rank=0 | step=98670/100000 progress=98.7% loss=0.032013 lr=4.835327e-09 elapsed=36:41:44 avg_step=1.34s recent_step=1.15s eta=00:29:40 +2026-08-26 16:37:29 | INFO | train rank=0 | step=98680/100000 progress=98.7% loss=0.036514 lr=4.762900e-09 elapsed=36:41:59 avg_step=1.34s recent_step=1.50s eta=00:29:27 +2026-08-26 16:37:45 | INFO | train rank=0 | step=98690/100000 progress=98.7% loss=0.043678 lr=4.691019e-09 elapsed=36:42:15 avg_step=1.34s recent_step=1.56s eta=00:29:13 +2026-08-26 16:38:01 | INFO | train rank=0 | step=98700/100000 progress=98.7% loss=0.056213 lr=4.619685e-09 elapsed=36:42:31 avg_step=1.34s recent_step=1.63s eta=00:29:00 +2026-08-26 16:38:13 | INFO | train rank=0 | step=98710/100000 progress=98.7% loss=0.051971 lr=4.548897e-09 elapsed=36:42:43 avg_step=1.34s recent_step=1.18s eta=00:28:47 +2026-08-26 16:38:25 | INFO | train rank=0 | step=98720/100000 progress=98.7% loss=0.079163 lr=4.478655e-09 elapsed=36:42:55 avg_step=1.34s recent_step=1.17s eta=00:28:33 +2026-08-26 16:38:39 | INFO | train rank=0 | step=98730/100000 progress=98.7% loss=0.072205 lr=4.408960e-09 elapsed=36:43:09 avg_step=1.34s recent_step=1.44s eta=00:28:20 +2026-08-26 16:38:52 | INFO | train rank=0 | step=98740/100000 progress=98.7% loss=0.047119 lr=4.339811e-09 elapsed=36:43:22 avg_step=1.34s recent_step=1.27s eta=00:28:07 +2026-08-26 16:39:03 | INFO | train rank=0 | step=98750/100000 progress=98.8% loss=0.033012 lr=4.271208e-09 elapsed=36:43:33 avg_step=1.34s recent_step=1.12s eta=00:27:53 +2026-08-26 16:39:18 | INFO | train rank=0 | step=98760/100000 progress=98.8% loss=0.028008 lr=4.203152e-09 elapsed=36:43:48 avg_step=1.34s recent_step=1.49s eta=00:27:40 +2026-08-26 16:39:30 | INFO | train rank=0 | step=98770/100000 progress=98.8% loss=0.049500 lr=4.135642e-09 elapsed=36:44:00 avg_step=1.34s recent_step=1.20s eta=00:27:26 +2026-08-26 16:39:44 | INFO | train rank=0 | step=98780/100000 progress=98.8% loss=0.046890 lr=4.068678e-09 elapsed=36:44:14 avg_step=1.34s recent_step=1.44s eta=00:27:13 +2026-08-26 16:40:01 | INFO | train rank=0 | step=98790/100000 progress=98.8% loss=0.034866 lr=4.002260e-09 elapsed=36:44:31 avg_step=1.34s recent_step=1.63s eta=00:27:00 +2026-08-26 16:40:14 | INFO | train rank=0 | step=98800/100000 progress=98.8% loss=0.032471 lr=3.936389e-09 elapsed=36:44:44 avg_step=1.34s recent_step=1.34s eta=00:26:46 +2026-08-26 16:40:26 | INFO | train rank=0 | step=98810/100000 progress=98.8% loss=0.045349 lr=3.871065e-09 elapsed=36:44:55 avg_step=1.34s recent_step=1.15s eta=00:26:33 +2026-08-26 16:40:37 | INFO | train rank=0 | step=98820/100000 progress=98.8% loss=0.036346 lr=3.806286e-09 elapsed=36:45:07 avg_step=1.34s recent_step=1.17s eta=00:26:19 +2026-08-26 16:40:52 | INFO | train rank=0 | step=98830/100000 progress=98.8% loss=0.061188 lr=3.742054e-09 elapsed=36:45:22 avg_step=1.34s recent_step=1.50s eta=00:26:06 +2026-08-26 16:41:04 | INFO | train rank=0 | step=98840/100000 progress=98.8% loss=0.026237 lr=3.678369e-09 elapsed=36:45:34 avg_step=1.34s recent_step=1.19s eta=00:25:53 +2026-08-26 16:41:23 | INFO | train rank=0 | step=98850/100000 progress=98.9% loss=0.066833 lr=3.615230e-09 elapsed=36:45:53 avg_step=1.34s recent_step=1.88s eta=00:25:39 +2026-08-26 16:41:34 | INFO | train rank=0 | step=98860/100000 progress=98.9% loss=0.045471 lr=3.552637e-09 elapsed=36:46:04 avg_step=1.34s recent_step=1.14s eta=00:25:26 +2026-08-26 16:41:50 | INFO | train rank=0 | step=98870/100000 progress=98.9% loss=0.047089 lr=3.490590e-09 elapsed=36:46:20 avg_step=1.34s recent_step=1.54s eta=00:25:12 +2026-08-26 16:42:01 | INFO | train rank=0 | step=98880/100000 progress=98.9% loss=0.046761 lr=3.429090e-09 elapsed=36:46:31 avg_step=1.34s recent_step=1.16s eta=00:24:59 +2026-08-26 16:42:14 | INFO | train rank=0 | step=98890/100000 progress=98.9% loss=0.030899 lr=3.368137e-09 elapsed=36:46:44 avg_step=1.34s recent_step=1.26s eta=00:24:46 +2026-08-26 16:42:29 | INFO | train rank=0 | step=98900/100000 progress=98.9% loss=0.080963 lr=3.307730e-09 elapsed=36:46:59 avg_step=1.34s recent_step=1.48s eta=00:24:32 +2026-08-26 16:42:41 | INFO | train rank=0 | step=98910/100000 progress=98.9% loss=0.047104 lr=3.247869e-09 elapsed=36:47:11 avg_step=1.34s recent_step=1.24s eta=00:24:19 +2026-08-26 16:42:55 | INFO | train rank=0 | step=98920/100000 progress=98.9% loss=0.036560 lr=3.188555e-09 elapsed=36:47:25 avg_step=1.34s recent_step=1.37s eta=00:24:06 +2026-08-26 16:43:10 | INFO | train rank=0 | step=98930/100000 progress=98.9% loss=0.039841 lr=3.129787e-09 elapsed=36:47:39 avg_step=1.34s recent_step=1.48s eta=00:23:52 +2026-08-26 16:43:21 | INFO | train rank=0 | step=98940/100000 progress=98.9% loss=0.026901 lr=3.071566e-09 elapsed=36:47:51 avg_step=1.34s recent_step=1.16s eta=00:23:39 +2026-08-26 16:43:34 | INFO | train rank=0 | step=98950/100000 progress=99.0% loss=0.023308 lr=3.013891e-09 elapsed=36:48:04 avg_step=1.34s recent_step=1.26s eta=00:23:25 +2026-08-26 16:43:48 | INFO | train rank=0 | step=98960/100000 progress=99.0% loss=0.043503 lr=2.956762e-09 elapsed=36:48:18 avg_step=1.34s recent_step=1.39s eta=00:23:12 +2026-08-26 16:44:01 | INFO | train rank=0 | step=98970/100000 progress=99.0% loss=0.057327 lr=2.900180e-09 elapsed=36:48:31 avg_step=1.34s recent_step=1.37s eta=00:22:59 +2026-08-26 16:44:13 | INFO | train rank=0 | step=98980/100000 progress=99.0% loss=0.038551 lr=2.844145e-09 elapsed=36:48:43 avg_step=1.34s recent_step=1.16s eta=00:22:45 +2026-08-26 16:44:28 | INFO | train rank=0 | step=98990/100000 progress=99.0% loss=0.041153 lr=2.788656e-09 elapsed=36:48:58 avg_step=1.34s recent_step=1.52s eta=00:22:32 +2026-08-26 16:44:40 | INFO | train rank=0 | step=99000/100000 progress=99.0% loss=0.034157 lr=2.733713e-09 elapsed=36:49:10 avg_step=1.34s recent_step=1.20s eta=00:22:18 +2026-08-26 16:44:55 | INFO | train rank=0 | step=99010/100000 progress=99.0% loss=0.041389 lr=2.679317e-09 elapsed=36:49:25 avg_step=1.34s recent_step=1.46s eta=00:22:05 +2026-08-26 16:45:07 | INFO | train rank=0 | step=99020/100000 progress=99.0% loss=0.058281 lr=2.625468e-09 elapsed=36:49:37 avg_step=1.34s recent_step=1.19s eta=00:21:52 +2026-08-26 16:45:20 | INFO | train rank=0 | step=99030/100000 progress=99.0% loss=0.063004 lr=2.572165e-09 elapsed=36:49:50 avg_step=1.34s recent_step=1.36s eta=00:21:38 +2026-08-26 16:45:32 | INFO | train rank=0 | step=99040/100000 progress=99.0% loss=0.047668 lr=2.519408e-09 elapsed=36:50:02 avg_step=1.34s recent_step=1.19s eta=00:21:25 +2026-08-26 16:45:44 | INFO | train rank=0 | step=99050/100000 progress=99.1% loss=0.073578 lr=2.467198e-09 elapsed=36:50:14 avg_step=1.34s recent_step=1.14s eta=00:21:11 +2026-08-26 16:45:57 | INFO | train rank=0 | step=99060/100000 progress=99.1% loss=0.049072 lr=2.415535e-09 elapsed=36:50:27 avg_step=1.34s recent_step=1.32s eta=00:20:58 +2026-08-26 16:46:12 | INFO | train rank=0 | step=99070/100000 progress=99.1% loss=0.060822 lr=2.364418e-09 elapsed=36:50:42 avg_step=1.34s recent_step=1.53s eta=00:20:45 +2026-08-26 16:46:24 | INFO | train rank=0 | step=99080/100000 progress=99.1% loss=0.027878 lr=2.313847e-09 elapsed=36:50:54 avg_step=1.34s recent_step=1.22s eta=00:20:31 +2026-08-26 16:46:41 | INFO | train rank=0 | step=99090/100000 progress=99.1% loss=0.030136 lr=2.263823e-09 elapsed=36:51:11 avg_step=1.34s recent_step=1.64s eta=00:20:18 +2026-08-26 16:46:58 | INFO | train rank=0 | step=99100/100000 progress=99.1% loss=0.042465 lr=2.214346e-09 elapsed=36:51:28 avg_step=1.34s recent_step=1.68s eta=00:20:05 +2026-08-26 16:47:11 | INFO | train rank=0 | step=99110/100000 progress=99.1% loss=0.050079 lr=2.165415e-09 elapsed=36:51:41 avg_step=1.34s recent_step=1.36s eta=00:19:51 +2026-08-26 16:47:23 | INFO | train rank=0 | step=99120/100000 progress=99.1% loss=0.078583 lr=2.117031e-09 elapsed=36:51:53 avg_step=1.34s recent_step=1.14s eta=00:19:38 +2026-08-26 16:47:35 | INFO | train rank=0 | step=99130/100000 progress=99.1% loss=0.034958 lr=2.069193e-09 elapsed=36:52:05 avg_step=1.34s recent_step=1.22s eta=00:19:24 +2026-08-26 16:47:49 | INFO | train rank=0 | step=99140/100000 progress=99.1% loss=0.046066 lr=2.021902e-09 elapsed=36:52:19 avg_step=1.34s recent_step=1.41s eta=00:19:11 +2026-08-26 16:48:01 | INFO | train rank=0 | step=99150/100000 progress=99.2% loss=0.049164 lr=1.975158e-09 elapsed=36:52:31 avg_step=1.34s recent_step=1.18s eta=00:18:58 +2026-08-26 16:48:16 | INFO | train rank=0 | step=99160/100000 progress=99.2% loss=0.042511 lr=1.928960e-09 elapsed=36:52:46 avg_step=1.34s recent_step=1.54s eta=00:18:44 +2026-08-26 16:48:28 | INFO | train rank=0 | step=99170/100000 progress=99.2% loss=0.058060 lr=1.883308e-09 elapsed=36:52:57 avg_step=1.34s recent_step=1.14s eta=00:18:31 +2026-08-26 16:48:41 | INFO | train rank=0 | step=99180/100000 progress=99.2% loss=0.051392 lr=1.838204e-09 elapsed=36:53:11 avg_step=1.34s recent_step=1.37s eta=00:18:17 +2026-08-26 16:48:54 | INFO | train rank=0 | step=99190/100000 progress=99.2% loss=0.035004 lr=1.793646e-09 elapsed=36:53:24 avg_step=1.34s recent_step=1.30s eta=00:18:04 +2026-08-26 16:49:09 | INFO | train rank=0 | step=99200/100000 progress=99.2% loss=0.091431 lr=1.749634e-09 elapsed=36:53:39 avg_step=1.34s recent_step=1.46s eta=00:17:51 +2026-08-26 16:49:22 | INFO | train rank=0 | step=99210/100000 progress=99.2% loss=0.019848 lr=1.706169e-09 elapsed=36:53:52 avg_step=1.34s recent_step=1.30s eta=00:17:37 +2026-08-26 16:49:34 | INFO | train rank=0 | step=99220/100000 progress=99.2% loss=0.039948 lr=1.663251e-09 elapsed=36:54:04 avg_step=1.34s recent_step=1.21s eta=00:17:24 +2026-08-26 16:49:49 | INFO | train rank=0 | step=99230/100000 progress=99.2% loss=0.050125 lr=1.620879e-09 elapsed=36:54:18 avg_step=1.34s recent_step=1.46s eta=00:17:10 +2026-08-26 16:50:05 | INFO | train rank=0 | step=99240/100000 progress=99.2% loss=0.067139 lr=1.579054e-09 elapsed=36:54:35 avg_step=1.34s recent_step=1.67s eta=00:16:57 +2026-08-26 16:50:17 | INFO | train rank=0 | step=99250/100000 progress=99.2% loss=0.035286 lr=1.537775e-09 elapsed=36:54:46 avg_step=1.34s recent_step=1.13s eta=00:16:44 +2026-08-26 16:50:32 | INFO | train rank=0 | step=99260/100000 progress=99.3% loss=0.063690 lr=1.497043e-09 elapsed=36:55:02 avg_step=1.34s recent_step=1.59s eta=00:16:30 +2026-08-26 16:50:47 | INFO | train rank=0 | step=99270/100000 progress=99.3% loss=0.046310 lr=1.456858e-09 elapsed=36:55:17 avg_step=1.34s recent_step=1.46s eta=00:16:17 +2026-08-26 16:50:59 | INFO | train rank=0 | step=99280/100000 progress=99.3% loss=0.083313 lr=1.417219e-09 elapsed=36:55:29 avg_step=1.34s recent_step=1.18s eta=00:16:04 +2026-08-26 16:51:12 | INFO | train rank=0 | step=99290/100000 progress=99.3% loss=0.050919 lr=1.378127e-09 elapsed=36:55:42 avg_step=1.34s recent_step=1.35s eta=00:15:50 +2026-08-26 16:51:27 | INFO | train rank=0 | step=99300/100000 progress=99.3% loss=0.060242 lr=1.339582e-09 elapsed=36:55:56 avg_step=1.34s recent_step=1.42s eta=00:15:37 +2026-08-26 16:51:42 | INFO | train rank=0 | step=99310/100000 progress=99.3% loss=0.037018 lr=1.301583e-09 elapsed=36:56:12 avg_step=1.34s recent_step=1.52s eta=00:15:23 +2026-08-26 16:51:58 | INFO | train rank=0 | step=99320/100000 progress=99.3% loss=0.023407 lr=1.264131e-09 elapsed=36:56:28 avg_step=1.34s recent_step=1.64s eta=00:15:10 +2026-08-26 16:52:10 | INFO | train rank=0 | step=99330/100000 progress=99.3% loss=0.052231 lr=1.227226e-09 elapsed=36:56:40 avg_step=1.34s recent_step=1.17s eta=00:14:57 +2026-08-26 16:52:21 | INFO | train rank=0 | step=99340/100000 progress=99.3% loss=0.040863 lr=1.190867e-09 elapsed=36:56:51 avg_step=1.34s recent_step=1.14s eta=00:14:43 +2026-08-26 16:52:33 | INFO | train rank=0 | step=99350/100000 progress=99.4% loss=0.038040 lr=1.155055e-09 elapsed=36:57:03 avg_step=1.34s recent_step=1.14s eta=00:14:30 +2026-08-26 16:52:44 | INFO | train rank=0 | step=99360/100000 progress=99.4% loss=0.043381 lr=1.119789e-09 elapsed=36:57:14 avg_step=1.34s recent_step=1.18s eta=00:14:16 +2026-08-26 16:52:56 | INFO | train rank=0 | step=99370/100000 progress=99.4% loss=0.024193 lr=1.085070e-09 elapsed=36:57:26 avg_step=1.34s recent_step=1.14s eta=00:14:03 +2026-08-26 16:53:07 | INFO | train rank=0 | step=99380/100000 progress=99.4% loss=0.050537 lr=1.050898e-09 elapsed=36:57:37 avg_step=1.34s recent_step=1.15s eta=00:13:50 +2026-08-26 16:53:22 | INFO | train rank=0 | step=99390/100000 progress=99.4% loss=0.054810 lr=1.017273e-09 elapsed=36:57:52 avg_step=1.34s recent_step=1.50s eta=00:13:36 +2026-08-26 16:53:42 | INFO | train rank=0 | step=99400/100000 progress=99.4% loss=0.045212 lr=9.841942e-10 elapsed=36:58:12 avg_step=1.34s recent_step=1.95s eta=00:13:23 +2026-08-26 16:53:56 | INFO | train rank=0 | step=99410/100000 progress=99.4% loss=0.052338 lr=9.516621e-10 elapsed=36:58:26 avg_step=1.34s recent_step=1.44s eta=00:13:09 +2026-08-26 16:54:08 | INFO | train rank=0 | step=99420/100000 progress=99.4% loss=0.019497 lr=9.196768e-10 elapsed=36:58:38 avg_step=1.34s recent_step=1.17s eta=00:12:56 +2026-08-26 16:54:20 | INFO | train rank=0 | step=99430/100000 progress=99.4% loss=0.058807 lr=8.882381e-10 elapsed=36:58:49 avg_step=1.34s recent_step=1.17s eta=00:12:43 +2026-08-26 16:54:36 | INFO | train rank=0 | step=99440/100000 progress=99.4% loss=0.027405 lr=8.573461e-10 elapsed=36:59:05 avg_step=1.34s recent_step=1.59s eta=00:12:29 +2026-08-26 16:54:47 | INFO | train rank=0 | step=99450/100000 progress=99.5% loss=0.030640 lr=8.270008e-10 elapsed=36:59:17 avg_step=1.34s recent_step=1.16s eta=00:12:16 +2026-08-26 16:54:59 | INFO | train rank=0 | step=99460/100000 progress=99.5% loss=0.051086 lr=7.972023e-10 elapsed=36:59:29 avg_step=1.34s recent_step=1.16s eta=00:12:03 +2026-08-26 16:55:10 | INFO | train rank=0 | step=99470/100000 progress=99.5% loss=0.051712 lr=7.679504e-10 elapsed=36:59:40 avg_step=1.34s recent_step=1.17s eta=00:11:49 +2026-08-26 16:55:22 | INFO | train rank=0 | step=99480/100000 progress=99.5% loss=0.046967 lr=7.392452e-10 elapsed=36:59:52 avg_step=1.34s recent_step=1.16s eta=00:11:36 +2026-08-26 16:55:37 | INFO | train rank=0 | step=99490/100000 progress=99.5% loss=0.031239 lr=7.110868e-10 elapsed=37:00:07 avg_step=1.34s recent_step=1.47s eta=00:11:22 +2026-08-26 16:55:51 | INFO | train rank=0 | step=99500/100000 progress=99.5% loss=0.044479 lr=6.834750e-10 elapsed=37:00:21 avg_step=1.34s recent_step=1.42s eta=00:11:09 +2026-08-26 16:56:02 | INFO | train rank=0 | step=99510/100000 progress=99.5% loss=0.023560 lr=6.564100e-10 elapsed=37:00:32 avg_step=1.34s recent_step=1.14s eta=00:10:56 +2026-08-26 16:56:22 | INFO | train rank=0 | step=99520/100000 progress=99.5% loss=0.045258 lr=6.298917e-10 elapsed=37:00:52 avg_step=1.34s recent_step=1.93s eta=00:10:42 +2026-08-26 16:56:33 | INFO | train rank=0 | step=99530/100000 progress=99.5% loss=0.048569 lr=6.039201e-10 elapsed=37:01:03 avg_step=1.34s recent_step=1.15s eta=00:10:29 +2026-08-26 16:56:45 | INFO | train rank=0 | step=99540/100000 progress=99.5% loss=0.036209 lr=5.784953e-10 elapsed=37:01:15 avg_step=1.34s recent_step=1.17s eta=00:10:15 +2026-08-26 16:57:01 | INFO | train rank=0 | step=99550/100000 progress=99.6% loss=0.042923 lr=5.536172e-10 elapsed=37:01:31 avg_step=1.34s recent_step=1.61s eta=00:10:02 +2026-08-26 16:57:14 | INFO | train rank=0 | step=99560/100000 progress=99.6% loss=0.036346 lr=5.292858e-10 elapsed=37:01:44 avg_step=1.34s recent_step=1.34s eta=00:09:49 +2026-08-26 16:57:31 | INFO | train rank=0 | step=99570/100000 progress=99.6% loss=0.065216 lr=5.055011e-10 elapsed=37:02:01 avg_step=1.34s recent_step=1.67s eta=00:09:35 +2026-08-26 16:57:45 | INFO | train rank=0 | step=99580/100000 progress=99.6% loss=0.056763 lr=4.822632e-10 elapsed=37:02:15 avg_step=1.34s recent_step=1.40s eta=00:09:22 +2026-08-26 16:57:56 | INFO | train rank=0 | step=99590/100000 progress=99.6% loss=0.039658 lr=4.595720e-10 elapsed=37:02:26 avg_step=1.34s recent_step=1.12s eta=00:09:08 +2026-08-26 16:58:08 | INFO | train rank=0 | step=99600/100000 progress=99.6% loss=0.040031 lr=4.374276e-10 elapsed=37:02:38 avg_step=1.34s recent_step=1.17s eta=00:08:55 +2026-08-26 16:58:20 | INFO | train rank=0 | step=99610/100000 progress=99.6% loss=0.051041 lr=4.158299e-10 elapsed=37:02:49 avg_step=1.34s recent_step=1.16s eta=00:08:42 +2026-08-26 16:58:32 | INFO | train rank=0 | step=99620/100000 progress=99.6% loss=0.060120 lr=3.947790e-10 elapsed=37:03:02 avg_step=1.34s recent_step=1.21s eta=00:08:28 +2026-08-26 16:58:46 | INFO | train rank=0 | step=99630/100000 progress=99.6% loss=0.062256 lr=3.742748e-10 elapsed=37:03:16 avg_step=1.34s recent_step=1.44s eta=00:08:15 +2026-08-26 16:58:58 | INFO | train rank=0 | step=99640/100000 progress=99.6% loss=0.041737 lr=3.543173e-10 elapsed=37:03:28 avg_step=1.34s recent_step=1.19s eta=00:08:02 +2026-08-26 16:59:12 | INFO | train rank=0 | step=99650/100000 progress=99.7% loss=0.046318 lr=3.349067e-10 elapsed=37:03:42 avg_step=1.34s recent_step=1.44s eta=00:07:48 +2026-08-26 16:59:26 | INFO | train rank=0 | step=99660/100000 progress=99.7% loss=0.026566 lr=3.160427e-10 elapsed=37:03:56 avg_step=1.34s recent_step=1.33s eta=00:07:35 +2026-08-26 16:59:42 | INFO | train rank=0 | step=99670/100000 progress=99.7% loss=0.036133 lr=2.977256e-10 elapsed=37:04:12 avg_step=1.34s recent_step=1.66s eta=00:07:21 +2026-08-26 16:59:54 | INFO | train rank=0 | step=99680/100000 progress=99.7% loss=0.039597 lr=2.799551e-10 elapsed=37:04:24 avg_step=1.34s recent_step=1.15s eta=00:07:08 +2026-08-26 17:00:08 | INFO | train rank=0 | step=99690/100000 progress=99.7% loss=0.043831 lr=2.627315e-10 elapsed=37:04:38 avg_step=1.34s recent_step=1.42s eta=00:06:55 +2026-08-26 17:00:27 | INFO | train rank=0 | step=99700/100000 progress=99.7% loss=0.024582 lr=2.460546e-10 elapsed=37:04:57 avg_step=1.34s recent_step=1.88s eta=00:06:41 +2026-08-26 17:00:39 | INFO | train rank=0 | step=99710/100000 progress=99.7% loss=0.033249 lr=2.299245e-10 elapsed=37:05:08 avg_step=1.34s recent_step=1.17s eta=00:06:28 +2026-08-26 17:00:53 | INFO | train rank=0 | step=99720/100000 progress=99.7% loss=0.031990 lr=2.143411e-10 elapsed=37:05:23 avg_step=1.34s recent_step=1.48s eta=00:06:14 +2026-08-26 17:01:08 | INFO | train rank=0 | step=99730/100000 progress=99.7% loss=0.034882 lr=1.993045e-10 elapsed=37:05:38 avg_step=1.34s recent_step=1.45s eta=00:06:01 +2026-08-26 17:01:20 | INFO | train rank=0 | step=99740/100000 progress=99.7% loss=0.042953 lr=1.848147e-10 elapsed=37:05:49 avg_step=1.34s recent_step=1.17s eta=00:05:48 +2026-08-26 17:01:31 | INFO | train rank=0 | step=99750/100000 progress=99.8% loss=0.078735 lr=1.708717e-10 elapsed=37:06:01 avg_step=1.34s recent_step=1.19s eta=00:05:34 +2026-08-26 17:01:43 | INFO | train rank=0 | step=99760/100000 progress=99.8% loss=0.052917 lr=1.574754e-10 elapsed=37:06:13 avg_step=1.34s recent_step=1.14s eta=00:05:21 +2026-08-26 17:01:57 | INFO | train rank=0 | step=99770/100000 progress=99.8% loss=0.066376 lr=1.446259e-10 elapsed=37:06:27 avg_step=1.34s recent_step=1.45s eta=00:05:07 +2026-08-26 17:02:09 | INFO | train rank=0 | step=99780/100000 progress=99.8% loss=0.034340 lr=1.323232e-10 elapsed=37:06:39 avg_step=1.34s recent_step=1.22s eta=00:04:54 +2026-08-26 17:02:24 | INFO | train rank=0 | step=99790/100000 progress=99.8% loss=0.030910 lr=1.205673e-10 elapsed=37:06:54 avg_step=1.34s recent_step=1.42s eta=00:04:41 +2026-08-26 17:02:38 | INFO | train rank=0 | step=99800/100000 progress=99.8% loss=0.037865 lr=1.093581e-10 elapsed=37:07:08 avg_step=1.34s recent_step=1.47s eta=00:04:27 +2026-08-26 17:02:50 | INFO | train rank=0 | step=99810/100000 progress=99.8% loss=0.021797 lr=9.869572e-11 elapsed=37:07:20 avg_step=1.34s recent_step=1.16s eta=00:04:14 +2026-08-26 17:03:02 | INFO | train rank=0 | step=99820/100000 progress=99.8% loss=0.054108 lr=8.858012e-11 elapsed=37:07:32 avg_step=1.34s recent_step=1.18s eta=00:04:01 +2026-08-26 17:03:17 | INFO | train rank=0 | step=99830/100000 progress=99.8% loss=0.105225 lr=7.901131e-11 elapsed=37:07:47 avg_step=1.34s recent_step=1.50s eta=00:03:47 +2026-08-26 17:03:33 | INFO | train rank=0 | step=99840/100000 progress=99.8% loss=0.037849 lr=6.998928e-11 elapsed=37:08:03 avg_step=1.34s recent_step=1.59s eta=00:03:34 +2026-08-26 17:03:48 | INFO | train rank=0 | step=99850/100000 progress=99.9% loss=0.037811 lr=6.151403e-11 elapsed=37:08:18 avg_step=1.34s recent_step=1.52s eta=00:03:20 +2026-08-26 17:04:01 | INFO | train rank=0 | step=99860/100000 progress=99.9% loss=0.044601 lr=5.358557e-11 elapsed=37:08:31 avg_step=1.34s recent_step=1.34s eta=00:03:07 +2026-08-26 17:04:14 | INFO | train rank=0 | step=99870/100000 progress=99.9% loss=0.048172 lr=4.620389e-11 elapsed=37:08:44 avg_step=1.34s recent_step=1.25s eta=00:02:54 +2026-08-26 17:04:31 | INFO | train rank=0 | step=99880/100000 progress=99.9% loss=0.047386 lr=3.936901e-11 elapsed=37:09:01 avg_step=1.34s recent_step=1.71s eta=00:02:40 +2026-08-26 17:04:43 | INFO | train rank=0 | step=99890/100000 progress=99.9% loss=0.038437 lr=3.308091e-11 elapsed=37:09:12 avg_step=1.34s recent_step=1.15s eta=00:02:27 +2026-08-26 17:04:54 | INFO | train rank=0 | step=99900/100000 progress=99.9% loss=0.028259 lr=2.733960e-11 elapsed=37:09:24 avg_step=1.34s recent_step=1.16s eta=00:02:13 +2026-08-26 17:05:08 | INFO | train rank=0 | step=99910/100000 progress=99.9% loss=0.035568 lr=2.214508e-11 elapsed=37:09:38 avg_step=1.34s recent_step=1.43s eta=00:02:00 +2026-08-26 17:05:20 | INFO | train rank=0 | step=99920/100000 progress=99.9% loss=0.019089 lr=1.749735e-11 elapsed=37:09:50 avg_step=1.34s recent_step=1.17s eta=00:01:47 +2026-08-26 17:05:36 | INFO | train rank=0 | step=99930/100000 progress=99.9% loss=0.043518 lr=1.339641e-11 elapsed=37:10:06 avg_step=1.34s recent_step=1.63s eta=00:01:33 +2026-08-26 17:05:48 | INFO | train rank=0 | step=99940/100000 progress=99.9% loss=0.022758 lr=9.842262e-12 elapsed=37:10:18 avg_step=1.34s recent_step=1.16s eta=00:01:20 +2026-08-26 17:06:03 | INFO | train rank=0 | step=99950/100000 progress=100.0% loss=0.059814 lr=6.834905e-12 elapsed=37:10:33 avg_step=1.34s recent_step=1.53s eta=00:01:06 +2026-08-26 17:06:17 | INFO | train rank=0 | step=99960/100000 progress=100.0% loss=0.046051 lr=4.374339e-12 elapsed=37:10:47 avg_step=1.34s recent_step=1.34s eta=00:00:53 +2026-08-26 17:06:28 | INFO | train rank=0 | step=99970/100000 progress=100.0% loss=0.066132 lr=2.460566e-12 elapsed=37:10:58 avg_step=1.34s recent_step=1.17s eta=00:00:40 +2026-08-26 17:06:41 | INFO | train rank=0 | step=99980/100000 progress=100.0% loss=0.052597 lr=1.093585e-12 elapsed=37:11:11 avg_step=1.34s recent_step=1.24s eta=00:00:26 +2026-08-26 17:07:00 | INFO | train rank=0 | step=99990/100000 progress=100.0% loss=0.054962 lr=2.733962e-13 elapsed=37:11:30 avg_step=1.34s recent_step=1.92s eta=00:00:13 +2026-08-26 17:07:11 | INFO | train rank=0 | step=100000/100000 progress=100.0% loss=0.022404 lr=0.000000e+00 elapsed=37:11:41 avg_step=1.34s recent_step=1.13s eta=00:00:00 +2026-08-26 17:07:38 | INFO | train rank=0 | Saved DeepSpeed checkpoint to /data/NTU_slab/draven/checkpoints/codebase/codebase_v7_droid/steps100k_bs128_lr1e-5_qwen3vl2b_new_policy/ds_checkpoint_100000 +2026-08-26 17:07:38 | INFO | train rank=0 | Finished training final_step=100000/100000 elapsed=37:12:08 avg_step=1.34s +2026-08-26 17:07:38 | INFO | train rank=0 | Training finished +2026-08-26 17:08:02 | INFO | train rank=0 | Saved final DeepSpeed checkpoint to /data/NTU_slab/draven/checkpoints/codebase/codebase_v7_droid/steps100k_bs128_lr1e-5_qwen3vl2b_new_policy/ds_checkpoint_final