Download K12/run/train.log from SlayerLab/dowody400m-checkpoints: direct link, hf CLI and curl.
- Browser
- Download file 14.3 kB
-
https://huggingface.co/SlayerLab/dowody400m-checkpoints/resolve/main/K12/run/train.log
- Command line
-
hf download hf://SlayerLab/dowody400m-checkpoints/K12/run/train.log
-
curl -L -o train.log https://huggingface.co/SlayerLab/dowody400m-checkpoints/resolve/main/K12/run/train.log
14.3 kB
| [04:33:11] device=cuda:0 bf16=True dev=NVIDIA GeForce RTX 4090 world=1 | |
| [04:33:11] dane: train=2,500,001,211B block=4096 batch=128 tok/step=524,288 | |
| [04:33:13] model GPT-ref: 184.6M param (L12 d1024 h16) | |
| [04:33:14] optimizer=muon muon_lr=0.02 (mult=33.3x) | |
| [04:33:14] r11 pamiec: act_ckpt=True loss_chunk=1024 | |
| [04:33:14] r13: mixer_layout=AAAAAAAAAAAA gdn_head_dim=64 attn_bias=0 rope_style=interleaved | |
| [04:33:15] torch.compile enabled | |
| [04:33:15] START gpt-ref: steps=4768 (od 0) lr=0.0006->6e-05 | |
| [04:37:14] step 20/4768 loss 10.4046 lr 2.40e-05 gnorm 2.87 43,863 tok/s peakVRAM 14.2GB | |
| [04:40:12] step 40/4768 loss 9.5343 lr 4.80e-05 gnorm 1.38 58,709 tok/s peakVRAM 14.2GB | |
| [04:43:11] step 60/4768 loss 8.5928 lr 7.20e-05 gnorm 0.59 58,729 tok/s peakVRAM 14.2GB | |
| [04:46:09] step 80/4768 loss 7.9727 lr 9.60e-05 gnorm 0.86 58,711 tok/s peakVRAM 14.2GB | |
| [04:49:08] step 100/4768 loss 7.5379 lr 1.20e-04 gnorm 0.89 58,683 tok/s peakVRAM 14.2GB | |
| [04:52:07] step 120/4768 loss 7.0513 lr 1.44e-04 gnorm 0.87 58,663 tok/s peakVRAM 14.2GB | |
| [04:55:06] step 140/4768 loss 6.6284 lr 1.68e-04 gnorm 0.71 58,646 tok/s peakVRAM 14.2GB | |
| [04:58:05] step 160/4768 loss 6.2768 lr 1.92e-04 gnorm 0.77 58,628 tok/s peakVRAM 14.2GB | |
| [05:01:03] step 180/4768 loss 6.0184 lr 2.16e-04 gnorm 0.77 58,626 tok/s peakVRAM 14.2GB | |
| [05:04:02] step 200/4768 loss 5.8071 lr 2.40e-04 gnorm 0.83 58,612 tok/s peakVRAM 14.2GB | |
| [05:07:01] step 220/4768 loss 5.6286 lr 2.64e-04 gnorm 0.80 58,612 tok/s peakVRAM 14.2GB | |
| [05:10:00] step 240/4768 loss 5.4628 lr 2.88e-04 gnorm 0.86 58,623 tok/s peakVRAM 14.2GB | |
| [05:12:59] step 260/4768 loss 5.3007 lr 3.12e-04 gnorm 0.86 58,626 tok/s peakVRAM 14.2GB | |
| [05:15:58] step 280/4768 loss 5.1497 lr 3.36e-04 gnorm 0.74 58,625 tok/s peakVRAM 14.2GB | |
| [05:18:57] step 300/4768 loss 5.0103 lr 3.60e-04 gnorm 0.83 58,621 tok/s peakVRAM 14.2GB | |
| [05:21:56] step 320/4768 loss 4.8812 lr 3.84e-04 gnorm 0.59 58,614 tok/s peakVRAM 14.2GB | |
| [05:24:54] step 340/4768 loss 4.7429 lr 4.08e-04 gnorm 0.76 58,616 tok/s peakVRAM 14.2GB | |
| [05:27:53] step 360/4768 loss 4.6169 lr 4.32e-04 gnorm 0.55 58,614 tok/s peakVRAM 14.2GB | |
| [05:30:52] step 380/4768 loss 4.4968 lr 4.56e-04 gnorm 0.53 58,616 tok/s peakVRAM 14.2GB | |
| [05:33:51] step 400/4768 loss 4.3926 lr 4.80e-04 gnorm 0.41 58,609 tok/s peakVRAM 14.2GB | |
| [05:36:50] step 420/4768 loss 4.3094 lr 5.04e-04 gnorm 0.46 58,611 tok/s peakVRAM 14.2GB | |
| [05:39:49] step 440/4768 loss 4.2352 lr 5.28e-04 gnorm 0.44 58,612 tok/s peakVRAM 14.2GB | |
| [05:42:48] step 460/4768 loss 4.1637 lr 5.52e-04 gnorm 0.39 58,612 tok/s peakVRAM 14.2GB | |
| [05:45:47] step 480/4768 loss 4.1191 lr 5.76e-04 gnorm 0.34 58,613 tok/s peakVRAM 14.2GB | |
| [05:48:46] step 500/4768 loss 4.0618 lr 6.00e-04 gnorm 0.32 58,609 tok/s peakVRAM 14.2GB | |
| [05:48:47] ckpt @ 500 | |
| [05:51:46] step 520/4768 loss 4.0278 lr 6.00e-04 gnorm 0.33 58,065 tok/s peakVRAM 14.2GB | |
| [05:54:45] step 540/4768 loss 3.9765 lr 6.00e-04 gnorm 0.25 58,620 tok/s peakVRAM 14.2GB | |
| [05:57:44] step 560/4768 loss 3.9295 lr 6.00e-04 gnorm 0.26 58,621 tok/s peakVRAM 14.2GB | |
| [06:00:43] step 580/4768 loss 3.8875 lr 6.00e-04 gnorm 0.26 58,617 tok/s peakVRAM 14.2GB | |
| [06:03:42] step 600/4768 loss 3.8484 lr 6.00e-04 gnorm 0.21 58,617 tok/s peakVRAM 14.2GB | |
| [06:06:41] step 620/4768 loss 3.8276 lr 6.00e-04 gnorm 0.23 58,610 tok/s peakVRAM 14.2GB | |
| [06:09:40] step 640/4768 loss 3.8017 lr 6.00e-04 gnorm 0.18 58,612 tok/s peakVRAM 14.2GB | |
| [06:12:39] step 660/4768 loss 3.7725 lr 6.00e-04 gnorm 0.19 58,615 tok/s peakVRAM 14.2GB | |
| [06:15:37] step 680/4768 loss 3.7515 lr 6.00e-04 gnorm 0.19 58,607 tok/s peakVRAM 14.2GB | |
| [06:18:36] step 700/4768 loss 3.7217 lr 6.00e-04 gnorm 0.18 58,603 tok/s peakVRAM 14.2GB | |
| [06:21:35] step 720/4768 loss 3.6930 lr 6.00e-04 gnorm 0.18 58,599 tok/s peakVRAM 14.2GB | |
| [06:24:34] step 740/4768 loss 3.6890 lr 6.00e-04 gnorm 0.19 58,600 tok/s peakVRAM 14.2GB | |
| [06:27:33] step 760/4768 loss 3.6650 lr 6.00e-04 gnorm 0.16 58,596 tok/s peakVRAM 14.2GB | |
| [06:30:32] step 780/4768 loss 3.6402 lr 6.00e-04 gnorm 0.17 58,597 tok/s peakVRAM 14.2GB | |
| [06:33:31] step 800/4768 loss 3.6266 lr 6.00e-04 gnorm 0.16 58,601 tok/s peakVRAM 14.2GB | |
| [06:36:30] step 820/4768 loss 3.6111 lr 6.00e-04 gnorm 0.16 58,599 tok/s peakVRAM 14.2GB | |
| [06:39:29] step 840/4768 loss 3.5983 lr 6.00e-04 gnorm 0.16 58,600 tok/s peakVRAM 14.2GB | |
| [06:42:28] step 860/4768 loss 3.5771 lr 6.00e-04 gnorm 0.16 58,598 tok/s peakVRAM 14.2GB | |
| [06:45:27] step 880/4768 loss 3.5720 lr 6.00e-04 gnorm 0.15 58,592 tok/s peakVRAM 14.2GB | |
| [06:48:26] step 900/4768 loss 3.5545 lr 6.00e-04 gnorm 0.15 58,586 tok/s peakVRAM 14.2GB | |
| [06:51:25] step 920/4768 loss 3.5473 lr 6.00e-04 gnorm 0.14 58,593 tok/s peakVRAM 14.2GB | |
| [06:54:24] step 940/4768 loss 3.5419 lr 6.00e-04 gnorm 0.16 58,587 tok/s peakVRAM 14.2GB | |
| [06:57:23] step 960/4768 loss 3.5175 lr 6.00e-04 gnorm 0.14 58,589 tok/s peakVRAM 14.2GB | |
| [07:00:22] step 980/4768 loss 3.5143 lr 6.00e-04 gnorm 0.14 58,595 tok/s peakVRAM 14.2GB | |
| [07:03:21] step 1000/4768 loss 3.5043 lr 6.00e-04 gnorm 0.15 58,588 tok/s peakVRAM 14.2GB | |
| [07:03:23] ckpt @ 1000 | |
| [07:06:22] step 1020/4768 loss 3.4993 lr 6.00e-04 gnorm 0.14 57,812 tok/s peakVRAM 14.2GB | |
| [07:09:21] step 1040/4768 loss 3.4728 lr 6.00e-04 gnorm 0.13 58,591 tok/s peakVRAM 14.2GB | |
| [07:12:20] step 1060/4768 loss 3.4845 lr 6.00e-04 gnorm 0.14 58,588 tok/s peakVRAM 14.2GB | |
| [07:15:19] step 1080/4768 loss 3.4739 lr 6.00e-04 gnorm 0.16 58,587 tok/s peakVRAM 14.2GB | |
| [07:18:18] step 1100/4768 loss 3.4584 lr 6.00e-04 gnorm 0.15 58,592 tok/s peakVRAM 14.2GB | |
| [07:21:17] step 1120/4768 loss 3.4405 lr 6.00e-04 gnorm 0.12 58,595 tok/s peakVRAM 14.2GB | |
| [07:24:16] step 1140/4768 loss 3.4338 lr 6.00e-04 gnorm 0.14 58,588 tok/s peakVRAM 14.2GB | |
| [07:27:15] step 1160/4768 loss 3.4284 lr 6.00e-04 gnorm 0.12 58,585 tok/s peakVRAM 14.2GB | |
| [07:30:14] step 1180/4768 loss 3.4317 lr 6.00e-04 gnorm 0.14 58,590 tok/s peakVRAM 14.2GB | |
| [07:33:13] step 1200/4768 loss 3.4130 lr 6.00e-04 gnorm 0.14 58,581 tok/s peakVRAM 14.2GB | |
| [07:36:12] step 1220/4768 loss 3.4098 lr 6.00e-04 gnorm 0.12 58,584 tok/s peakVRAM 14.2GB | |
| [07:39:11] step 1240/4768 loss 3.4026 lr 6.00e-04 gnorm 0.12 58,586 tok/s peakVRAM 14.2GB | |
| [07:42:10] step 1260/4768 loss 3.3888 lr 6.00e-04 gnorm 0.12 58,586 tok/s peakVRAM 14.2GB | |
| [07:45:09] step 1280/4768 loss 3.3914 lr 6.00e-04 gnorm 0.12 58,589 tok/s peakVRAM 14.2GB | |
| [07:48:08] step 1300/4768 loss 3.3844 lr 6.00e-04 gnorm 0.12 58,586 tok/s peakVRAM 14.2GB | |
| [07:51:07] step 1320/4768 loss 3.3767 lr 6.00e-04 gnorm 0.14 58,590 tok/s peakVRAM 14.2GB | |
| [07:54:06] step 1340/4768 loss 3.3768 lr 6.00e-04 gnorm 0.12 58,587 tok/s peakVRAM 14.2GB | |
| [07:57:05] step 1360/4768 loss 3.3681 lr 6.00e-04 gnorm 0.13 58,590 tok/s peakVRAM 14.2GB | |
| [08:00:04] step 1380/4768 loss 3.3594 lr 6.00e-04 gnorm 0.12 58,580 tok/s peakVRAM 14.2GB | |
| [08:03:03] step 1400/4768 loss 3.3493 lr 6.00e-04 gnorm 0.12 58,579 tok/s peakVRAM 14.2GB | |
| [08:06:02] step 1420/4768 loss 3.3370 lr 6.00e-04 gnorm 0.13 58,576 tok/s peakVRAM 14.2GB | |
| [08:09:01] step 1440/4768 loss 3.3438 lr 6.00e-04 gnorm 0.13 58,580 tok/s peakVRAM 14.2GB | |
| [08:12:00] step 1460/4768 loss 3.3417 lr 6.00e-04 gnorm 0.12 58,573 tok/s peakVRAM 14.2GB | |
| [08:14:59] step 1480/4768 loss 3.3458 lr 6.00e-04 gnorm 0.11 58,573 tok/s peakVRAM 14.2GB | |
| [08:17:58] step 1500/4768 loss 3.3185 lr 6.00e-04 gnorm 0.13 58,573 tok/s peakVRAM 14.2GB | |
| [08:18:00] ckpt @ 1500 | |
| [08:20:59] step 1520/4768 loss 3.3255 lr 6.00e-04 gnorm 0.11 57,778 tok/s peakVRAM 14.2GB | |
| [08:23:58] step 1540/4768 loss 3.3099 lr 6.00e-04 gnorm 0.12 58,532 tok/s peakVRAM 14.2GB | |
| [08:26:57] step 1560/4768 loss 3.3181 lr 6.00e-04 gnorm 0.12 58,532 tok/s peakVRAM 14.2GB | |
| [08:29:57] step 1580/4768 loss 3.3065 lr 6.00e-04 gnorm 0.13 58,567 tok/s peakVRAM 14.2GB | |
| [08:32:56] step 1600/4768 loss 3.2964 lr 6.00e-04 gnorm 0.12 58,572 tok/s peakVRAM 14.2GB | |
| [08:35:55] step 1620/4768 loss 3.2961 lr 6.00e-04 gnorm 0.13 58,576 tok/s peakVRAM 14.2GB | |
| [08:38:54] step 1640/4768 loss 3.2932 lr 6.00e-04 gnorm 0.11 58,575 tok/s peakVRAM 14.2GB | |
| [08:41:53] step 1660/4768 loss 3.2939 lr 6.00e-04 gnorm 0.12 58,565 tok/s peakVRAM 14.2GB | |
| [08:44:52] step 1680/4768 loss 3.2829 lr 6.00e-04 gnorm 0.11 58,562 tok/s peakVRAM 14.2GB | |
| [08:47:51] step 1700/4768 loss 3.2761 lr 6.00e-04 gnorm 0.13 58,570 tok/s peakVRAM 14.2GB | |
| [08:50:50] step 1720/4768 loss 3.2781 lr 6.00e-04 gnorm 0.12 58,566 tok/s peakVRAM 14.2GB | |
| [08:53:49] step 1740/4768 loss 3.2715 lr 6.00e-04 gnorm 0.11 58,556 tok/s peakVRAM 14.2GB | |
| [08:56:48] step 1760/4768 loss 3.2704 lr 6.00e-04 gnorm 0.13 58,551 tok/s peakVRAM 14.2GB | |
| [08:59:47] step 1780/4768 loss 3.2580 lr 6.00e-04 gnorm 0.11 58,559 tok/s peakVRAM 14.2GB | |
| [09:02:46] step 1800/4768 loss 3.2615 lr 6.00e-04 gnorm 0.12 58,560 tok/s peakVRAM 14.2GB | |
| [09:05:45] step 1820/4768 loss 3.2574 lr 6.00e-04 gnorm 0.12 58,560 tok/s peakVRAM 14.2GB | |
| [09:08:44] step 1840/4768 loss 3.2504 lr 6.00e-04 gnorm 0.11 58,555 tok/s peakVRAM 14.2GB | |
| [09:11:43] step 1860/4768 loss 3.2495 lr 6.00e-04 gnorm 0.12 58,571 tok/s peakVRAM 14.2GB | |
| [09:14:42] step 1880/4768 loss 3.2457 lr 6.00e-04 gnorm 0.11 58,554 tok/s peakVRAM 14.2GB | |
| [09:17:41] step 1900/4768 loss 3.2398 lr 6.00e-04 gnorm 0.12 58,562 tok/s peakVRAM 14.2GB | |
| [09:20:40] step 1920/4768 loss 3.2400 lr 6.00e-04 gnorm 0.11 58,556 tok/s peakVRAM 14.2GB | |
| [09:23:39] step 1940/4768 loss 3.2326 lr 6.00e-04 gnorm 0.11 58,557 tok/s peakVRAM 14.2GB | |
| [09:26:39] step 1960/4768 loss 3.2364 lr 6.00e-04 gnorm 0.12 58,559 tok/s peakVRAM 14.2GB | |
| [09:29:38] step 1980/4768 loss 3.2304 lr 6.00e-04 gnorm 0.11 58,575 tok/s peakVRAM 14.2GB | |
| [09:32:37] step 2000/4768 loss 3.2241 lr 6.00e-04 gnorm 0.11 58,571 tok/s peakVRAM 14.2GB | |
| [09:32:39] ckpt @ 2000 | |
| [09:35:38] step 2020/4768 loss 3.2240 lr 6.00e-04 gnorm 0.11 57,804 tok/s peakVRAM 14.2GB | |
| [09:38:37] step 2040/4768 loss 3.2166 lr 6.00e-04 gnorm 0.11 58,583 tok/s peakVRAM 14.2GB | |
| [09:41:36] step 2060/4768 loss 3.2176 lr 6.00e-04 gnorm 0.11 58,579 tok/s peakVRAM 14.2GB | |
| [09:44:35] step 2080/4768 loss 3.2063 lr 6.00e-04 gnorm 0.11 58,579 tok/s peakVRAM 14.2GB | |
| [09:47:34] step 2100/4768 loss 3.2144 lr 6.00e-04 gnorm 0.10 58,579 tok/s peakVRAM 14.2GB | |
| [09:50:33] step 2120/4768 loss 3.2037 lr 6.00e-04 gnorm 0.11 58,579 tok/s peakVRAM 14.2GB | |
| [09:53:32] step 2140/4768 loss 3.2170 lr 6.00e-04 gnorm 0.11 58,588 tok/s peakVRAM 14.2GB | |
| [09:56:31] step 2160/4768 loss 3.1952 lr 6.00e-04 gnorm 0.11 58,585 tok/s peakVRAM 14.2GB | |
| [09:59:30] step 2180/4768 loss 3.1976 lr 6.00e-04 gnorm 0.10 58,581 tok/s peakVRAM 14.2GB | |
| [10:02:29] step 2200/4768 loss 3.1985 lr 6.00e-04 gnorm 0.10 58,583 tok/s peakVRAM 14.2GB | |
| [10:05:28] step 2220/4768 loss 3.1905 lr 6.00e-04 gnorm 0.11 58,579 tok/s peakVRAM 14.2GB | |
| [10:08:27] step 2240/4768 loss 3.1876 lr 6.00e-04 gnorm 0.10 58,566 tok/s peakVRAM 14.2GB | |
| [10:11:26] step 2260/4768 loss 3.1840 lr 6.00e-04 gnorm 0.11 58,565 tok/s peakVRAM 14.2GB | |
| [10:14:25] step 2280/4768 loss 3.1804 lr 6.00e-04 gnorm 0.10 58,565 tok/s peakVRAM 14.2GB | |
| [10:17:24] step 2300/4768 loss 3.1842 lr 6.00e-04 gnorm 0.11 58,561 tok/s peakVRAM 14.2GB | |
| [10:20:23] step 2320/4768 loss 3.1801 lr 6.00e-04 gnorm 0.12 58,573 tok/s peakVRAM 14.2GB | |
| [10:23:22] step 2340/4768 loss 3.1748 lr 6.00e-04 gnorm 0.11 58,584 tok/s peakVRAM 14.2GB | |
| [10:26:21] step 2360/4768 loss 3.1654 lr 6.00e-04 gnorm 0.11 58,582 tok/s peakVRAM 14.2GB | |
| [10:29:20] step 2380/4768 loss 3.1644 lr 6.00e-04 gnorm 0.10 58,584 tok/s peakVRAM 14.2GB | |
| [10:32:19] step 2400/4768 loss 3.1648 lr 6.00e-04 gnorm 0.12 58,586 tok/s peakVRAM 14.2GB | |
| [10:35:18] step 2420/4768 loss 3.1676 lr 6.00e-04 gnorm 0.11 58,586 tok/s peakVRAM 14.2GB | |
| [10:38:17] step 2440/4768 loss 3.1628 lr 6.00e-04 gnorm 0.10 58,580 tok/s peakVRAM 14.2GB | |
| [10:41:16] step 2460/4768 loss 3.1619 lr 6.00e-04 gnorm 0.10 58,582 tok/s peakVRAM 14.2GB | |
| [10:44:15] step 2480/4768 loss 3.1585 lr 6.00e-04 gnorm 0.11 58,589 tok/s peakVRAM 14.2GB | |
| [10:47:14] step 2500/4768 loss 3.1526 lr 6.00e-04 gnorm 0.09 58,585 tok/s peakVRAM 14.2GB | |
| [10:47:16] ckpt @ 2500 | |
| [10:50:15] step 2520/4768 loss 3.1547 lr 6.00e-04 gnorm 0.12 57,830 tok/s peakVRAM 14.2GB | |
| [10:53:14] step 2540/4768 loss 3.1546 lr 6.00e-04 gnorm 0.11 58,617 tok/s peakVRAM 14.2GB | |
| [10:56:13] step 2560/4768 loss 3.1520 lr 6.00e-04 gnorm 0.11 58,598 tok/s peakVRAM 14.2GB | |
| [10:59:12] step 2580/4768 loss 3.1469 lr 6.00e-04 gnorm 0.10 58,573 tok/s peakVRAM 14.2GB | |
| [11:02:11] step 2600/4768 loss 3.1377 lr 6.00e-04 gnorm 0.11 58,616 tok/s peakVRAM 14.2GB | |
| [11:05:10] step 2620/4768 loss 3.1459 lr 6.00e-04 gnorm 0.11 58,586 tok/s peakVRAM 14.2GB | |
| [11:08:09] step 2640/4768 loss 3.1474 lr 6.00e-04 gnorm 0.11 58,584 tok/s peakVRAM 14.2GB | |
| [11:11:08] step 2660/4768 loss 3.1453 lr 6.00e-04 gnorm 0.11 58,589 tok/s peakVRAM 14.2GB | |
| [11:14:07] step 2680/4768 loss 3.1407 lr 6.00e-04 gnorm 0.10 58,578 tok/s peakVRAM 14.2GB | |
| [11:17:06] step 2700/4768 loss 3.1430 lr 6.00e-04 gnorm 0.11 58,575 tok/s peakVRAM 14.2GB | |
| [11:20:05] step 2720/4768 loss 3.1259 lr 6.00e-04 gnorm 0.10 58,573 tok/s peakVRAM 14.2GB | |
| [11:23:04] step 2740/4768 loss 3.1274 lr 6.00e-04 gnorm 0.11 58,564 tok/s peakVRAM 14.2GB | |
| [11:26:03] step 2760/4768 loss 3.1306 lr 6.00e-04 gnorm 0.11 58,568 tok/s peakVRAM 14.2GB | |
| [11:29:02] step 2780/4768 loss 3.1218 lr 6.00e-04 gnorm 0.11 58,567 tok/s peakVRAM 14.2GB | |
| [11:32:01] step 2800/4768 loss 3.1167 lr 6.00e-04 gnorm 0.11 58,560 tok/s peakVRAM 14.2GB | |
| [11:35:00] step 2820/4768 loss 3.1136 lr 6.00e-04 gnorm 0.10 58,564 tok/s peakVRAM 14.2GB | |
| [11:37:59] step 2840/4768 loss 3.1220 lr 6.00e-04 gnorm 0.10 58,585 tok/s peakVRAM 14.2GB | |
| [11:40:58] step 2860/4768 loss 3.1131 lr 6.00e-04 gnorm 0.10 58,602 tok/s peakVRAM 14.2GB | |
| [11:43:57] step 2880/4768 loss 3.1154 lr 6.00e-04 gnorm 0.10 58,558 tok/s peakVRAM 14.2GB | |
| [11:46:56] step 2900/4768 loss 3.1166 lr 6.00e-04 gnorm 0.10 58,605 tok/s peakVRAM 14.2GB | |
| [11:49:55] step 2920/4768 loss 3.1239 lr 6.00e-04 gnorm 0.10 58,617 tok/s peakVRAM 14.2GB | |
| [11:52:54] step 2940/4768 loss 3.1144 lr 6.00e-04 gnorm 0.09 58,618 tok/s peakVRAM 14.2GB | |
| [11:55:53] step 2960/4768 loss 3.1160 lr 6.00e-04 gnorm 0.11 58,609 tok/s peakVRAM 14.2GB | |
| [11:58:52] step 2980/4768 loss 3.1031 lr 6.00e-04 gnorm 0.10 58,602 tok/s peakVRAM 14.2GB | |
| [12:01:51] step 3000/4768 loss 3.1038 lr 6.00e-04 gnorm 0.09 58,570 tok/s peakVRAM 14.2GB | |
| [12:01:53] ckpt @ 3000 | |
| [12:04:52] step 3020/4768 loss 3.0994 lr 6.00e-04 gnorm 0.09 57,789 tok/s peakVRAM 14.2GB | |
| [12:07:51] step 3040/4768 loss 3.0966 lr 6.00e-04 gnorm 0.10 58,562 tok/s peakVRAM 14.2GB | |
| [12:10:50] step 3060/4768 loss 3.1002 lr 6.00e-04 gnorm 0.11 58,560 tok/s peakVRAM 14.2GB | |