cinimod-devops / full_val_eval.log
dkudos's picture
Upload full_val_eval.log with huggingface_hub
86ba7c5 verified
Raw
History Blame Contribute Delete
71.8 kB
[Eval] Loading checkpoint outputs/devops-300m-4096-bf16-vast/checkpoint-4000
[transformers] LlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
- If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
[Eval] 287.3M parameters loaded
[Dataset] 17492 files, 123,656 non-overlapping samples, seq_len=4096
[transformers] `use_return_dict` is deprecated! Use `return_dict` instead!
[Eval] batch 1 | samples 8/123656 | mean_loss 2.4351 | 1s elapsed | 35,948 tok/s
[Eval] batch 21 | samples 168/123656 | mean_loss 0.5549 | 9s elapsed | 75,776 tok/s
[Eval] batch 41 | samples 328/123656 | mean_loss 0.4678 | 17s elapsed | 77,938 tok/s
[Eval] batch 61 | samples 488/123656 | mean_loss 0.4119 | 25s elapsed | 78,609 tok/s
[Eval] batch 81 | samples 648/123656 | mean_loss 0.3887 | 34s elapsed | 78,920 tok/s
[Eval] batch 101 | samples 808/123656 | mean_loss 0.3729 | 42s elapsed | 79,039 tok/s
[Eval] batch 121 | samples 968/123656 | mean_loss 0.3641 | 50s elapsed | 79,081 tok/s
[Eval] batch 141 | samples 1128/123656 | mean_loss 0.3509 | 58s elapsed | 79,099 tok/s
[Eval] batch 161 | samples 1288/123656 | mean_loss 0.3459 | 67s elapsed | 79,107 tok/s
[Eval] batch 181 | samples 1448/123656 | mean_loss 0.3395 | 75s elapsed | 79,110 tok/s
[Eval] batch 201 | samples 1608/123656 | mean_loss 0.3343 | 83s elapsed | 79,102 tok/s
[Eval] batch 221 | samples 1768/123656 | mean_loss 0.3313 | 92s elapsed | 79,089 tok/s
[Eval] batch 241 | samples 1928/123656 | mean_loss 0.3274 | 100s elapsed | 79,048 tok/s
[Eval] batch 261 | samples 2088/123656 | mean_loss 0.4029 | 108s elapsed | 79,049 tok/s
[Eval] batch 281 | samples 2248/123656 | mean_loss 0.5424 | 117s elapsed | 79,037 tok/s
[Eval] batch 301 | samples 2408/123656 | mean_loss 0.6806 | 125s elapsed | 79,016 tok/s
[Eval] batch 321 | samples 2568/123656 | mean_loss 0.7765 | 133s elapsed | 79,013 tok/s
[Eval] batch 341 | samples 2728/123656 | mean_loss 0.8807 | 141s elapsed | 78,973 tok/s
[Eval] batch 361 | samples 2888/123656 | mean_loss 0.9719 | 150s elapsed | 78,961 tok/s
[Eval] batch 381 | samples 3048/123656 | mean_loss 1.0495 | 158s elapsed | 78,948 tok/s
[Eval] batch 401 | samples 3208/123656 | mean_loss 1.1496 | 166s elapsed | 78,949 tok/s
[Eval] batch 421 | samples 3368/123656 | mean_loss 1.2136 | 175s elapsed | 78,921 tok/s
[Eval] batch 441 | samples 3528/123656 | mean_loss 1.2720 | 183s elapsed | 78,920 tok/s
[Eval] batch 461 | samples 3688/123656 | mean_loss 1.3302 | 191s elapsed | 78,908 tok/s
[Eval] batch 481 | samples 3848/123656 | mean_loss 1.3822 | 200s elapsed | 78,904 tok/s
[Eval] batch 501 | samples 4008/123656 | mean_loss 1.4212 | 208s elapsed | 78,884 tok/s
[Eval] batch 521 | samples 4168/123656 | mean_loss 1.4664 | 216s elapsed | 78,861 tok/s
[Eval] batch 541 | samples 4328/123656 | mean_loss 1.5031 | 225s elapsed | 78,864 tok/s
[Eval] batch 561 | samples 4488/123656 | mean_loss 1.5387 | 233s elapsed | 78,841 tok/s
[Eval] batch 581 | samples 4648/123656 | mean_loss 1.5600 | 242s elapsed | 78,833 tok/s
[Eval] batch 601 | samples 4808/123656 | mean_loss 1.5791 | 250s elapsed | 78,831 tok/s
[Eval] batch 621 | samples 4968/123656 | mean_loss 1.6022 | 258s elapsed | 78,812 tok/s
[Eval] batch 641 | samples 5128/123656 | mean_loss 1.6348 | 266s elapsed | 78,816 tok/s
[Eval] batch 661 | samples 5288/123656 | mean_loss 1.6470 | 275s elapsed | 78,803 tok/s
[Eval] batch 681 | samples 5448/123656 | mean_loss 1.6770 | 283s elapsed | 78,806 tok/s
[Eval] batch 701 | samples 5608/123656 | mean_loss 1.6898 | 292s elapsed | 78,794 tok/s
[Eval] batch 721 | samples 5768/123656 | mean_loss 1.7145 | 300s elapsed | 78,776 tok/s
[Eval] batch 741 | samples 5928/123656 | mean_loss 1.7331 | 308s elapsed | 78,780 tok/s
[Eval] batch 761 | samples 6088/123656 | mean_loss 1.7482 | 317s elapsed | 78,783 tok/s
[Eval] batch 781 | samples 6248/123656 | mean_loss 1.7681 | 325s elapsed | 78,782 tok/s
[Eval] batch 801 | samples 6408/123656 | mean_loss 1.7910 | 333s elapsed | 78,779 tok/s
[Eval] batch 821 | samples 6568/123656 | mean_loss 1.8134 | 342s elapsed | 78,776 tok/s
[Eval] batch 841 | samples 6728/123656 | mean_loss 1.8247 | 350s elapsed | 78,767 tok/s
[Eval] batch 861 | samples 6888/123656 | mean_loss 1.8317 | 358s elapsed | 78,769 tok/s
[Eval] batch 881 | samples 7048/123656 | mean_loss 1.8502 | 367s elapsed | 78,760 tok/s
[Eval] batch 901 | samples 7208/123656 | mean_loss 1.8656 | 375s elapsed | 78,748 tok/s
[Eval] batch 921 | samples 7368/123656 | mean_loss 1.8812 | 383s elapsed | 78,750 tok/s
[Eval] batch 941 | samples 7528/123656 | mean_loss 1.8935 | 392s elapsed | 78,748 tok/s
[Eval] batch 961 | samples 7688/123656 | mean_loss 1.9068 | 400s elapsed | 78,751 tok/s
[Eval] batch 981 | samples 7848/123656 | mean_loss 1.9212 | 408s elapsed | 78,740 tok/s
[Eval] batch 1001 | samples 8008/123656 | mean_loss 1.9275 | 417s elapsed | 78,735 tok/s
[Eval] batch 1021 | samples 8168/123656 | mean_loss 1.9460 | 425s elapsed | 78,736 tok/s
[Eval] batch 1041 | samples 8328/123656 | mean_loss 1.9556 | 433s elapsed | 78,725 tok/s
[Eval] batch 1061 | samples 8488/123656 | mean_loss 1.9694 | 442s elapsed | 78,727 tok/s
[Eval] batch 1081 | samples 8648/123656 | mean_loss 1.9707 | 450s elapsed | 78,725 tok/s
[Eval] batch 1101 | samples 8808/123656 | mean_loss 1.9795 | 458s elapsed | 78,715 tok/s
[Eval] batch 1121 | samples 8968/123656 | mean_loss 1.9898 | 467s elapsed | 78,716 tok/s
[Eval] batch 1141 | samples 9128/123656 | mean_loss 1.9981 | 475s elapsed | 78,711 tok/s
[Eval] batch 1161 | samples 9288/123656 | mean_loss 2.0093 | 483s elapsed | 78,700 tok/s
[Eval] batch 1181 | samples 9448/123656 | mean_loss 2.0145 | 492s elapsed | 78,701 tok/s
[Eval] batch 1201 | samples 9608/123656 | mean_loss 2.0230 | 500s elapsed | 78,699 tok/s
[Eval] batch 1221 | samples 9768/123656 | mean_loss 2.0307 | 508s elapsed | 78,688 tok/s
[Eval] batch 1241 | samples 9928/123656 | mean_loss 2.0364 | 517s elapsed | 78,683 tok/s
[Eval] batch 1261 | samples 10088/123656 | mean_loss 2.0445 | 525s elapsed | 78,684 tok/s
[Eval] batch 1281 | samples 10248/123656 | mean_loss 2.0533 | 533s elapsed | 78,682 tok/s
[Eval] batch 1301 | samples 10408/123656 | mean_loss 2.0581 | 542s elapsed | 78,684 tok/s
[Eval] batch 1321 | samples 10568/123656 | mean_loss 2.0644 | 550s elapsed | 78,680 tok/s
[Eval] batch 1341 | samples 10728/123656 | mean_loss 2.0720 | 558s elapsed | 78,681 tok/s
[Eval] batch 1361 | samples 10888/123656 | mean_loss 2.0805 | 567s elapsed | 78,680 tok/s
[Eval] batch 1381 | samples 11048/123656 | mean_loss 2.0827 | 575s elapsed | 78,673 tok/s
[Eval] batch 1401 | samples 11208/123656 | mean_loss 2.0902 | 583s elapsed | 78,678 tok/s
[Eval] batch 1421 | samples 11368/123656 | mean_loss 2.0964 | 592s elapsed | 78,671 tok/s
[Eval] batch 1441 | samples 11528/123656 | mean_loss 2.1036 | 600s elapsed | 78,663 tok/s
[Eval] batch 1461 | samples 11688/123656 | mean_loss 2.1106 | 609s elapsed | 78,662 tok/s
[Eval] batch 1481 | samples 11848/123656 | mean_loss 2.1161 | 617s elapsed | 78,663 tok/s
[Eval] batch 1501 | samples 12008/123656 | mean_loss 2.1225 | 625s elapsed | 78,655 tok/s
[Eval] batch 1521 | samples 12168/123656 | mean_loss 2.1277 | 634s elapsed | 78,647 tok/s
[Eval] batch 1541 | samples 12328/123656 | mean_loss 2.1312 | 642s elapsed | 78,647 tok/s
[Eval] batch 1561 | samples 12488/123656 | mean_loss 2.1368 | 650s elapsed | 78,647 tok/s
[Eval] batch 1581 | samples 12648/123656 | mean_loss 2.1410 | 659s elapsed | 78,641 tok/s
[Eval] batch 1601 | samples 12808/123656 | mean_loss 2.1455 | 667s elapsed | 78,643 tok/s
[Eval] batch 1621 | samples 12968/123656 | mean_loss 2.1506 | 675s elapsed | 78,642 tok/s
[Eval] batch 1641 | samples 13128/123656 | mean_loss 2.1550 | 684s elapsed | 78,642 tok/s
[Eval] batch 1661 | samples 13288/123656 | mean_loss 2.1604 | 692s elapsed | 78,642 tok/s
[Eval] batch 1681 | samples 13448/123656 | mean_loss 2.1658 | 700s elapsed | 78,637 tok/s
[Eval] batch 1701 | samples 13608/123656 | mean_loss 2.1722 | 709s elapsed | 78,640 tok/s
[Eval] batch 1721 | samples 13768/123656 | mean_loss 2.1745 | 717s elapsed | 78,638 tok/s
[Eval] batch 1741 | samples 13928/123656 | mean_loss 2.1791 | 725s elapsed | 78,637 tok/s
[Eval] batch 1761 | samples 14088/123656 | mean_loss 2.1828 | 734s elapsed | 78,637 tok/s
[Eval] batch 1781 | samples 14248/123656 | mean_loss 2.1860 | 742s elapsed | 78,635 tok/s
[Eval] batch 1801 | samples 14408/123656 | mean_loss 2.1904 | 750s elapsed | 78,636 tok/s
[Eval] batch 1821 | samples 14568/123656 | mean_loss 2.1950 | 759s elapsed | 78,634 tok/s
[Eval] batch 1841 | samples 14728/123656 | mean_loss 2.1969 | 767s elapsed | 78,628 tok/s
[Eval] batch 1861 | samples 14888/123656 | mean_loss 2.2013 | 776s elapsed | 78,626 tok/s
[Eval] batch 1881 | samples 15048/123656 | mean_loss 2.2046 | 784s elapsed | 78,628 tok/s
[Eval] batch 1901 | samples 15208/123656 | mean_loss 2.2096 | 792s elapsed | 78,623 tok/s
[Eval] batch 1921 | samples 15368/123656 | mean_loss 2.2126 | 801s elapsed | 78,617 tok/s
[Eval] batch 1941 | samples 15528/123656 | mean_loss 2.2160 | 809s elapsed | 78,620 tok/s
[Eval] batch 1961 | samples 15688/123656 | mean_loss 2.2175 | 817s elapsed | 78,618 tok/s
[Eval] batch 1981 | samples 15848/123656 | mean_loss 2.2175 | 826s elapsed | 78,616 tok/s
[Eval] batch 2001 | samples 16008/123656 | mean_loss 2.2187 | 834s elapsed | 78,618 tok/s
[Eval] batch 2021 | samples 16168/123656 | mean_loss 2.2200 | 842s elapsed | 78,614 tok/s
[Eval] batch 2041 | samples 16328/123656 | mean_loss 2.2234 | 851s elapsed | 78,616 tok/s
[Eval] batch 2061 | samples 16488/123656 | mean_loss 2.2263 | 859s elapsed | 78,616 tok/s
[Eval] batch 2081 | samples 16648/123656 | mean_loss 2.2291 | 867s elapsed | 78,617 tok/s
[Eval] batch 2101 | samples 16808/123656 | mean_loss 2.2318 | 876s elapsed | 78,617 tok/s
[Eval] batch 2121 | samples 16968/123656 | mean_loss 2.2349 | 884s elapsed | 78,614 tok/s
[Eval] batch 2141 | samples 17128/123656 | mean_loss 2.2381 | 892s elapsed | 78,615 tok/s
[Eval] batch 2161 | samples 17288/123656 | mean_loss 2.2403 | 901s elapsed | 78,613 tok/s
[Eval] batch 2181 | samples 17448/123656 | mean_loss 2.2410 | 909s elapsed | 78,608 tok/s
[Eval] batch 2201 | samples 17608/123656 | mean_loss 2.2435 | 917s elapsed | 78,607 tok/s
[Eval] batch 2221 | samples 17768/123656 | mean_loss 2.2463 | 926s elapsed | 78,609 tok/s
[Eval] batch 2241 | samples 17928/123656 | mean_loss 2.2492 | 934s elapsed | 78,605 tok/s
[Eval] batch 2261 | samples 18088/123656 | mean_loss 2.2505 | 943s elapsed | 78,601 tok/s
[Eval] batch 2281 | samples 18248/123656 | mean_loss 2.2533 | 951s elapsed | 78,602 tok/s
[Eval] batch 2301 | samples 18408/123656 | mean_loss 2.2556 | 959s elapsed | 78,601 tok/s
[Eval] batch 2321 | samples 18568/123656 | mean_loss 2.2558 | 968s elapsed | 78,598 tok/s
[Eval] batch 2341 | samples 18728/123656 | mean_loss 2.2585 | 976s elapsed | 78,600 tok/s
[Eval] batch 2361 | samples 18888/123656 | mean_loss 2.2608 | 984s elapsed | 78,598 tok/s
[Eval] batch 2381 | samples 19048/123656 | mean_loss 2.2628 | 993s elapsed | 78,594 tok/s
[Eval] batch 2401 | samples 19208/123656 | mean_loss 2.2647 | 1001s elapsed | 78,594 tok/s
[Eval] batch 2421 | samples 19368/123656 | mean_loss 2.2670 | 1009s elapsed | 78,595 tok/s
[Eval] batch 2441 | samples 19528/123656 | mean_loss 2.2687 | 1018s elapsed | 78,591 tok/s
[Eval] batch 2461 | samples 19688/123656 | mean_loss 2.2713 | 1026s elapsed | 78,589 tok/s
[Eval] batch 2481 | samples 19848/123656 | mean_loss 2.2732 | 1034s elapsed | 78,591 tok/s
[Eval] batch 2501 | samples 20008/123656 | mean_loss 2.2744 | 1043s elapsed | 78,588 tok/s
[Eval] batch 2521 | samples 20168/123656 | mean_loss 2.2756 | 1051s elapsed | 78,587 tok/s
[Eval] batch 2541 | samples 20328/123656 | mean_loss 2.2810 | 1059s elapsed | 78,589 tok/s
[Eval] batch 2561 | samples 20488/123656 | mean_loss 2.2824 | 1068s elapsed | 78,585 tok/s
[Eval] batch 2581 | samples 20648/123656 | mean_loss 2.2848 | 1076s elapsed | 78,583 tok/s
[Eval] batch 2601 | samples 20808/123656 | mean_loss 2.2865 | 1085s elapsed | 78,586 tok/s
[Eval] batch 2621 | samples 20968/123656 | mean_loss 2.2866 | 1093s elapsed | 78,583 tok/s
[Eval] batch 2641 | samples 21128/123656 | mean_loss 2.2882 | 1101s elapsed | 78,581 tok/s
[Eval] batch 2661 | samples 21288/123656 | mean_loss 2.2896 | 1110s elapsed | 78,583 tok/s
[Eval] batch 2681 | samples 21448/123656 | mean_loss 2.2907 | 1118s elapsed | 78,581 tok/s
[Eval] batch 2701 | samples 21608/123656 | mean_loss 2.2958 | 1126s elapsed | 78,577 tok/s
[Eval] batch 2721 | samples 21768/123656 | mean_loss 2.3010 | 1135s elapsed | 78,578 tok/s
[Eval] batch 2741 | samples 21928/123656 | mean_loss 2.3032 | 1143s elapsed | 78,579 tok/s
[Eval] batch 2761 | samples 22088/123656 | mean_loss 2.3048 | 1151s elapsed | 78,577 tok/s
[Eval] batch 2781 | samples 22248/123656 | mean_loss 2.3010 | 1160s elapsed | 78,579 tok/s
[Eval] batch 2801 | samples 22408/123656 | mean_loss 2.3022 | 1168s elapsed | 78,577 tok/s
[Eval] batch 2821 | samples 22568/123656 | mean_loss 2.3038 | 1176s elapsed | 78,579 tok/s
[Eval] batch 2841 | samples 22728/123656 | mean_loss 2.3055 | 1185s elapsed | 78,578 tok/s
[Eval] batch 2861 | samples 22888/123656 | mean_loss 2.3052 | 1193s elapsed | 78,577 tok/s
[Eval] batch 2881 | samples 23048/123656 | mean_loss 2.3059 | 1201s elapsed | 78,579 tok/s
[Eval] batch 2901 | samples 23208/123656 | mean_loss 2.3081 | 1210s elapsed | 78,577 tok/s
[Eval] batch 2921 | samples 23368/123656 | mean_loss 2.3068 | 1218s elapsed | 78,578 tok/s
[Eval] batch 2941 | samples 23528/123656 | mean_loss 2.3083 | 1226s elapsed | 78,578 tok/s
[Eval] batch 2961 | samples 23688/123656 | mean_loss 2.3084 | 1235s elapsed | 78,574 tok/s
[Eval] batch 2981 | samples 23848/123656 | mean_loss 2.3096 | 1243s elapsed | 78,571 tok/s
[Eval] batch 3001 | samples 24008/123656 | mean_loss 2.3104 | 1252s elapsed | 78,572 tok/s
[Eval] batch 3021 | samples 24168/123656 | mean_loss 2.3127 | 1260s elapsed | 78,572 tok/s
[Eval] batch 3041 | samples 24328/123656 | mean_loss 2.3135 | 1268s elapsed | 78,569 tok/s
[Eval] batch 3061 | samples 24488/123656 | mean_loss 2.3155 | 1277s elapsed | 78,569 tok/s
[Eval] batch 3081 | samples 24648/123656 | mean_loss 2.3198 | 1285s elapsed | 78,570 tok/s
[Eval] batch 3101 | samples 24808/123656 | mean_loss 2.3205 | 1293s elapsed | 78,567 tok/s
[Eval] batch 3121 | samples 24968/123656 | mean_loss 2.3220 | 1302s elapsed | 78,568 tok/s
[Eval] batch 3141 | samples 25128/123656 | mean_loss 2.3240 | 1310s elapsed | 78,569 tok/s
[Eval] batch 3161 | samples 25288/123656 | mean_loss 2.3255 | 1318s elapsed | 78,565 tok/s
[Eval] batch 3181 | samples 25448/123656 | mean_loss 2.3270 | 1327s elapsed | 78,567 tok/s
[Eval] batch 3201 | samples 25608/123656 | mean_loss 2.3288 | 1335s elapsed | 78,567 tok/s
[Eval] batch 3221 | samples 25768/123656 | mean_loss 2.3291 | 1343s elapsed | 78,564 tok/s
[Eval] batch 3241 | samples 25928/123656 | mean_loss 2.3291 | 1352s elapsed | 78,561 tok/s
[Eval] batch 3261 | samples 26088/123656 | mean_loss 2.3304 | 1360s elapsed | 78,561 tok/s
[Eval] batch 3281 | samples 26248/123656 | mean_loss 2.3331 | 1368s elapsed | 78,562 tok/s
[Eval] batch 3301 | samples 26408/123656 | mean_loss 2.3337 | 1377s elapsed | 78,562 tok/s
[Eval] batch 3321 | samples 26568/123656 | mean_loss 2.3359 | 1385s elapsed | 78,563 tok/s
[Eval] batch 3341 | samples 26728/123656 | mean_loss 2.3376 | 1394s elapsed | 78,561 tok/s
[Eval] batch 3361 | samples 26888/123656 | mean_loss 2.3394 | 1402s elapsed | 78,560 tok/s
[Eval] batch 3381 | samples 27048/123656 | mean_loss 2.3391 | 1410s elapsed | 78,562 tok/s
[Eval] batch 3401 | samples 27208/123656 | mean_loss 2.3408 | 1419s elapsed | 78,561 tok/s
[Eval] batch 3421 | samples 27368/123656 | mean_loss 2.3410 | 1427s elapsed | 78,562 tok/s
[Eval] batch 3441 | samples 27528/123656 | mean_loss 2.3415 | 1435s elapsed | 78,561 tok/s
[Eval] batch 3461 | samples 27688/123656 | mean_loss 2.3423 | 1444s elapsed | 78,560 tok/s
[Eval] batch 3481 | samples 27848/123656 | mean_loss 2.3426 | 1452s elapsed | 78,561 tok/s
[Eval] batch 3501 | samples 28008/123656 | mean_loss 2.3422 | 1460s elapsed | 78,559 tok/s
[Eval] batch 3521 | samples 28168/123656 | mean_loss 2.3433 | 1469s elapsed | 78,558 tok/s
[Eval] batch 3541 | samples 28328/123656 | mean_loss 2.3442 | 1477s elapsed | 78,559 tok/s
[Eval] batch 3561 | samples 28488/123656 | mean_loss 2.3459 | 1485s elapsed | 78,558 tok/s
[Eval] batch 3581 | samples 28648/123656 | mean_loss 2.3466 | 1494s elapsed | 78,556 tok/s
[Eval] batch 3601 | samples 28808/123656 | mean_loss 2.3481 | 1502s elapsed | 78,558 tok/s
[Eval] batch 3621 | samples 28968/123656 | mean_loss 2.3526 | 1510s elapsed | 78,557 tok/s
[Eval] batch 3641 | samples 29128/123656 | mean_loss 2.3542 | 1519s elapsed | 78,557 tok/s
[Eval] batch 3661 | samples 29288/123656 | mean_loss 2.3552 | 1527s elapsed | 78,558 tok/s
[Eval] batch 3681 | samples 29448/123656 | mean_loss 2.3555 | 1535s elapsed | 78,558 tok/s
[Eval] batch 3701 | samples 29608/123656 | mean_loss 2.3547 | 1544s elapsed | 78,559 tok/s
[Eval] batch 3721 | samples 29768/123656 | mean_loss 2.3553 | 1552s elapsed | 78,557 tok/s
[Eval] batch 3741 | samples 29928/123656 | mean_loss 2.3558 | 1560s elapsed | 78,557 tok/s
[Eval] batch 3761 | samples 30088/123656 | mean_loss 2.3569 | 1569s elapsed | 78,558 tok/s
[Eval] batch 3781 | samples 30248/123656 | mean_loss 2.3569 | 1577s elapsed | 78,556 tok/s
[Eval] batch 3801 | samples 30408/123656 | mean_loss 2.3573 | 1586s elapsed | 78,554 tok/s
[Eval] batch 3821 | samples 30568/123656 | mean_loss 2.3593 | 1594s elapsed | 78,555 tok/s
[Eval] batch 3841 | samples 30728/123656 | mean_loss 2.3595 | 1602s elapsed | 78,555 tok/s
[Eval] batch 3861 | samples 30888/123656 | mean_loss 2.3600 | 1611s elapsed | 78,555 tok/s
[Eval] batch 3881 | samples 31048/123656 | mean_loss 2.3609 | 1619s elapsed | 78,556 tok/s
[Eval] batch 3901 | samples 31208/123656 | mean_loss 2.3601 | 1627s elapsed | 78,556 tok/s
[Eval] batch 3921 | samples 31368/123656 | mean_loss 2.3601 | 1636s elapsed | 78,557 tok/s
[Eval] batch 3941 | samples 31528/123656 | mean_loss 2.3587 | 1644s elapsed | 78,557 tok/s
[Eval] batch 3961 | samples 31688/123656 | mean_loss 2.3588 | 1652s elapsed | 78,558 tok/s
[Eval] batch 3981 | samples 31848/123656 | mean_loss 2.3601 | 1661s elapsed | 78,557 tok/s
[Eval] batch 4001 | samples 32008/123656 | mean_loss 2.3606 | 1669s elapsed | 78,554 tok/s
[Eval] batch 4021 | samples 32168/123656 | mean_loss 2.3608 | 1677s elapsed | 78,555 tok/s
[Eval] batch 4041 | samples 32328/123656 | mean_loss 2.3618 | 1686s elapsed | 78,555 tok/s
[Eval] batch 4061 | samples 32488/123656 | mean_loss 2.3626 | 1694s elapsed | 78,555 tok/s
[Eval] batch 4081 | samples 32648/123656 | mean_loss 2.3648 | 1702s elapsed | 78,556 tok/s
[Eval] batch 4101 | samples 32808/123656 | mean_loss 2.3657 | 1711s elapsed | 78,556 tok/s
[Eval] batch 4121 | samples 32968/123656 | mean_loss 2.3665 | 1719s elapsed | 78,557 tok/s
[Eval] batch 4141 | samples 33128/123656 | mean_loss 2.3672 | 1727s elapsed | 78,556 tok/s
[Eval] batch 4161 | samples 33288/123656 | mean_loss 2.3671 | 1736s elapsed | 78,556 tok/s
[Eval] batch 4181 | samples 33448/123656 | mean_loss 2.3675 | 1744s elapsed | 78,557 tok/s
[Eval] batch 4201 | samples 33608/123656 | mean_loss 2.3678 | 1752s elapsed | 78,554 tok/s
[Eval] batch 4221 | samples 33768/123656 | mean_loss 2.3680 | 1761s elapsed | 78,553 tok/s
[Eval] batch 4241 | samples 33928/123656 | mean_loss 2.3735 | 1769s elapsed | 78,555 tok/s
[Eval] batch 4261 | samples 34088/123656 | mean_loss 2.3742 | 1777s elapsed | 78,554 tok/s
[Eval] batch 4281 | samples 34248/123656 | mean_loss 2.3763 | 1786s elapsed | 78,556 tok/s
[Eval] batch 4301 | samples 34408/123656 | mean_loss 2.3768 | 1794s elapsed | 78,555 tok/s
[Eval] batch 4321 | samples 34568/123656 | mean_loss 2.3773 | 1802s elapsed | 78,556 tok/s
[Eval] batch 4341 | samples 34728/123656 | mean_loss 2.3770 | 1811s elapsed | 78,556 tok/s
[Eval] batch 4361 | samples 34888/123656 | mean_loss 2.3776 | 1819s elapsed | 78,557 tok/s
[Eval] batch 4381 | samples 35048/123656 | mean_loss 2.3770 | 1827s elapsed | 78,557 tok/s
[Eval] batch 4401 | samples 35208/123656 | mean_loss 2.3754 | 1836s elapsed | 78,555 tok/s
[Eval] batch 4421 | samples 35368/123656 | mean_loss 2.3755 | 1844s elapsed | 78,556 tok/s
[Eval] batch 4441 | samples 35528/123656 | mean_loss 2.3760 | 1852s elapsed | 78,556 tok/s
[Eval] batch 4461 | samples 35688/123656 | mean_loss 2.3832 | 1861s elapsed | 78,556 tok/s
[Eval] batch 4481 | samples 35848/123656 | mean_loss 2.3874 | 1869s elapsed | 78,557 tok/s
[Eval] batch 4501 | samples 36008/123656 | mean_loss 2.3853 | 1878s elapsed | 78,555 tok/s
[Eval] batch 4521 | samples 36168/123656 | mean_loss 2.3842 | 1886s elapsed | 78,555 tok/s
[Eval] batch 4541 | samples 36328/123656 | mean_loss 2.3857 | 1894s elapsed | 78,556 tok/s
[Eval] batch 4561 | samples 36488/123656 | mean_loss 2.3850 | 1903s elapsed | 78,554 tok/s
[Eval] batch 4581 | samples 36648/123656 | mean_loss 2.3846 | 1911s elapsed | 78,553 tok/s
[Eval] batch 4601 | samples 36808/123656 | mean_loss 2.3846 | 1919s elapsed | 78,554 tok/s
[Eval] batch 4621 | samples 36968/123656 | mean_loss 2.3844 | 1928s elapsed | 78,553 tok/s
[Eval] batch 4641 | samples 37128/123656 | mean_loss 2.3839 | 1936s elapsed | 78,554 tok/s
[Eval] batch 4661 | samples 37288/123656 | mean_loss 2.3831 | 1944s elapsed | 78,555 tok/s
[Eval] batch 4681 | samples 37448/123656 | mean_loss 2.3832 | 1953s elapsed | 78,555 tok/s
[Eval] batch 4701 | samples 37608/123656 | mean_loss 2.3842 | 1961s elapsed | 78,556 tok/s
[Eval] batch 4721 | samples 37768/123656 | mean_loss 2.3834 | 1969s elapsed | 78,555 tok/s
[Eval] batch 4741 | samples 37928/123656 | mean_loss 2.3835 | 1978s elapsed | 78,556 tok/s
[Eval] batch 4761 | samples 38088/123656 | mean_loss 2.3830 | 1986s elapsed | 78,555 tok/s
[Eval] batch 4781 | samples 38248/123656 | mean_loss 2.3827 | 1994s elapsed | 78,556 tok/s
[Eval] batch 4801 | samples 38408/123656 | mean_loss 2.3815 | 2003s elapsed | 78,556 tok/s
[Eval] batch 4821 | samples 38568/123656 | mean_loss 2.3810 | 2011s elapsed | 78,557 tok/s
[Eval] batch 4841 | samples 38728/123656 | mean_loss 2.3807 | 2019s elapsed | 78,557 tok/s
[Eval] batch 4861 | samples 38888/123656 | mean_loss 2.3808 | 2028s elapsed | 78,555 tok/s
[Eval] batch 4881 | samples 39048/123656 | mean_loss 2.3806 | 2036s elapsed | 78,556 tok/s
[Eval] batch 4901 | samples 39208/123656 | mean_loss 2.3802 | 2044s elapsed | 78,556 tok/s
[Eval] batch 4921 | samples 39368/123656 | mean_loss 2.3798 | 2053s elapsed | 78,557 tok/s
[Eval] batch 4941 | samples 39528/123656 | mean_loss 2.3797 | 2061s elapsed | 78,558 tok/s
[Eval] batch 4961 | samples 39688/123656 | mean_loss 2.3792 | 2069s elapsed | 78,558 tok/s
[Eval] batch 4981 | samples 39848/123656 | mean_loss 2.3789 | 2078s elapsed | 78,559 tok/s
[Eval] batch 5001 | samples 40008/123656 | mean_loss 2.3792 | 2086s elapsed | 78,559 tok/s
[Eval] batch 5021 | samples 40168/123656 | mean_loss 2.3793 | 2094s elapsed | 78,559 tok/s
[Eval] batch 5041 | samples 40328/123656 | mean_loss 2.3788 | 2103s elapsed | 78,558 tok/s
[Eval] batch 5061 | samples 40488/123656 | mean_loss 2.3782 | 2111s elapsed | 78,556 tok/s
[Eval] batch 5081 | samples 40648/123656 | mean_loss 2.3777 | 2119s elapsed | 78,557 tok/s
[Eval] batch 5101 | samples 40808/123656 | mean_loss 2.3776 | 2128s elapsed | 78,557 tok/s
[Eval] batch 5121 | samples 40968/123656 | mean_loss 2.3776 | 2136s elapsed | 78,556 tok/s
[Eval] batch 5141 | samples 41128/123656 | mean_loss 2.3771 | 2144s elapsed | 78,556 tok/s
[Eval] batch 5161 | samples 41288/123656 | mean_loss 2.3765 | 2153s elapsed | 78,557 tok/s
[Eval] batch 5181 | samples 41448/123656 | mean_loss 2.3761 | 2161s elapsed | 78,558 tok/s
[Eval] batch 5201 | samples 41608/123656 | mean_loss 2.3767 | 2169s elapsed | 78,557 tok/s
[Eval] batch 5221 | samples 41768/123656 | mean_loss 2.3784 | 2178s elapsed | 78,556 tok/s
[Eval] batch 5241 | samples 41928/123656 | mean_loss 2.3781 | 2186s elapsed | 78,557 tok/s
[Eval] batch 5261 | samples 42088/123656 | mean_loss 2.3769 | 2195s elapsed | 78,556 tok/s
[Eval] batch 5281 | samples 42248/123656 | mean_loss 2.3766 | 2203s elapsed | 78,554 tok/s
[Eval] batch 5301 | samples 42408/123656 | mean_loss 2.3751 | 2211s elapsed | 78,555 tok/s
[Eval] batch 5321 | samples 42568/123656 | mean_loss 2.3745 | 2220s elapsed | 78,555 tok/s
[Eval] batch 5341 | samples 42728/123656 | mean_loss 2.3743 | 2228s elapsed | 78,554 tok/s
[Eval] batch 5361 | samples 42888/123656 | mean_loss 2.3734 | 2236s elapsed | 78,555 tok/s
[Eval] batch 5381 | samples 43048/123656 | mean_loss 2.3733 | 2245s elapsed | 78,554 tok/s
[Eval] batch 5401 | samples 43208/123656 | mean_loss 2.3732 | 2253s elapsed | 78,553 tok/s
[Eval] batch 5421 | samples 43368/123656 | mean_loss 2.3725 | 2261s elapsed | 78,553 tok/s
[Eval] batch 5441 | samples 43528/123656 | mean_loss 2.3716 | 2270s elapsed | 78,554 tok/s
[Eval] batch 5461 | samples 43688/123656 | mean_loss 2.3709 | 2278s elapsed | 78,553 tok/s
[Eval] batch 5481 | samples 43848/123656 | mean_loss 2.3707 | 2286s elapsed | 78,554 tok/s
[Eval] batch 5501 | samples 44008/123656 | mean_loss 2.3701 | 2295s elapsed | 78,553 tok/s
[Eval] batch 5521 | samples 44168/123656 | mean_loss 2.3693 | 2303s elapsed | 78,551 tok/s
[Eval] batch 5541 | samples 44328/123656 | mean_loss 2.3688 | 2311s elapsed | 78,552 tok/s
[Eval] batch 5561 | samples 44488/123656 | mean_loss 2.3682 | 2320s elapsed | 78,552 tok/s
[Eval] batch 5581 | samples 44648/123656 | mean_loss 2.3677 | 2328s elapsed | 78,553 tok/s
[Eval] batch 5601 | samples 44808/123656 | mean_loss 2.3674 | 2336s elapsed | 78,553 tok/s
[Eval] batch 5621 | samples 44968/123656 | mean_loss 2.3671 | 2345s elapsed | 78,551 tok/s
[Eval] batch 5641 | samples 45128/123656 | mean_loss 2.3668 | 2353s elapsed | 78,553 tok/s
[Eval] batch 5661 | samples 45288/123656 | mean_loss 2.3667 | 2361s elapsed | 78,552 tok/s
[Eval] batch 5681 | samples 45448/123656 | mean_loss 2.3669 | 2370s elapsed | 78,550 tok/s
[Eval] batch 5701 | samples 45608/123656 | mean_loss 2.3667 | 2378s elapsed | 78,551 tok/s
[Eval] batch 5721 | samples 45768/123656 | mean_loss 2.3670 | 2387s elapsed | 78,551 tok/s
[Eval] batch 5741 | samples 45928/123656 | mean_loss 2.3666 | 2395s elapsed | 78,550 tok/s
[Eval] batch 5761 | samples 46088/123656 | mean_loss 2.3665 | 2403s elapsed | 78,552 tok/s
[Eval] batch 5781 | samples 46248/123656 | mean_loss 2.3663 | 2412s elapsed | 78,551 tok/s
[Eval] batch 5801 | samples 46408/123656 | mean_loss 2.3665 | 2420s elapsed | 78,551 tok/s
[Eval] batch 5821 | samples 46568/123656 | mean_loss 2.3665 | 2428s elapsed | 78,552 tok/s
[Eval] batch 5841 | samples 46728/123656 | mean_loss 2.3665 | 2437s elapsed | 78,551 tok/s
[Eval] batch 5861 | samples 46888/123656 | mean_loss 2.3666 | 2445s elapsed | 78,552 tok/s
[Eval] batch 5881 | samples 47048/123656 | mean_loss 2.3664 | 2453s elapsed | 78,551 tok/s
[Eval] batch 5901 | samples 47208/123656 | mean_loss 2.3657 | 2462s elapsed | 78,550 tok/s
[Eval] batch 5921 | samples 47368/123656 | mean_loss 2.3652 | 2470s elapsed | 78,551 tok/s
[Eval] batch 5941 | samples 47528/123656 | mean_loss 2.3650 | 2478s elapsed | 78,550 tok/s
[Eval] batch 5961 | samples 47688/123656 | mean_loss 2.3647 | 2487s elapsed | 78,550 tok/s
[Eval] batch 5981 | samples 47848/123656 | mean_loss 2.3642 | 2495s elapsed | 78,551 tok/s
[Eval] batch 6001 | samples 48008/123656 | mean_loss 2.3643 | 2503s elapsed | 78,550 tok/s
[Eval] batch 6021 | samples 48168/123656 | mean_loss 2.3636 | 2512s elapsed | 78,551 tok/s
[Eval] batch 6041 | samples 48328/123656 | mean_loss 2.3633 | 2520s elapsed | 78,551 tok/s
[Eval] batch 6061 | samples 48488/123656 | mean_loss 2.3625 | 2528s elapsed | 78,551 tok/s
[Eval] batch 6081 | samples 48648/123656 | mean_loss 2.3625 | 2537s elapsed | 78,552 tok/s
[Eval] batch 6101 | samples 48808/123656 | mean_loss 2.3619 | 2545s elapsed | 78,551 tok/s
[Eval] batch 6121 | samples 48968/123656 | mean_loss 2.3616 | 2553s elapsed | 78,552 tok/s
[Eval] batch 6141 | samples 49128/123656 | mean_loss 2.3614 | 2562s elapsed | 78,551 tok/s
[Eval] batch 6161 | samples 49288/123656 | mean_loss 2.3609 | 2570s elapsed | 78,552 tok/s
[Eval] batch 6181 | samples 49448/123656 | mean_loss 2.3610 | 2578s elapsed | 78,552 tok/s
[Eval] batch 6201 | samples 49608/123656 | mean_loss 2.3610 | 2587s elapsed | 78,553 tok/s
[Eval] batch 6221 | samples 49768/123656 | mean_loss 2.3605 | 2595s elapsed | 78,553 tok/s
[Eval] batch 6241 | samples 49928/123656 | mean_loss 2.3593 | 2603s elapsed | 78,551 tok/s
[Eval] batch 6261 | samples 50088/123656 | mean_loss 2.3590 | 2612s elapsed | 78,550 tok/s
[Eval] batch 6281 | samples 50248/123656 | mean_loss 2.3586 | 2620s elapsed | 78,550 tok/s
[Eval] batch 6301 | samples 50408/123656 | mean_loss 2.3579 | 2629s elapsed | 78,551 tok/s
[Eval] batch 6321 | samples 50568/123656 | mean_loss 2.3575 | 2637s elapsed | 78,549 tok/s
[Eval] batch 6341 | samples 50728/123656 | mean_loss 2.3573 | 2645s elapsed | 78,549 tok/s
[Eval] batch 6361 | samples 50888/123656 | mean_loss 2.3570 | 2654s elapsed | 78,550 tok/s
[Eval] batch 6381 | samples 51048/123656 | mean_loss 2.3568 | 2662s elapsed | 78,548 tok/s
[Eval] batch 6401 | samples 51208/123656 | mean_loss 2.3568 | 2670s elapsed | 78,548 tok/s
[Eval] batch 6421 | samples 51368/123656 | mean_loss 2.3562 | 2679s elapsed | 78,549 tok/s
[Eval] batch 6441 | samples 51528/123656 | mean_loss 2.3555 | 2687s elapsed | 78,550 tok/s
[Eval] batch 6461 | samples 51688/123656 | mean_loss 2.3547 | 2695s elapsed | 78,550 tok/s
[Eval] batch 6481 | samples 51848/123656 | mean_loss 2.3542 | 2704s elapsed | 78,550 tok/s
[Eval] batch 6501 | samples 52008/123656 | mean_loss 2.3538 | 2712s elapsed | 78,551 tok/s
[Eval] batch 6521 | samples 52168/123656 | mean_loss 2.3537 | 2720s elapsed | 78,550 tok/s
[Eval] batch 6541 | samples 52328/123656 | mean_loss 2.3540 | 2729s elapsed | 78,550 tok/s
[Eval] batch 6561 | samples 52488/123656 | mean_loss 2.3536 | 2737s elapsed | 78,550 tok/s
[Eval] batch 6581 | samples 52648/123656 | mean_loss 2.3537 | 2745s elapsed | 78,549 tok/s
[Eval] batch 6601 | samples 52808/123656 | mean_loss 2.3532 | 2754s elapsed | 78,550 tok/s
[Eval] batch 6621 | samples 52968/123656 | mean_loss 2.3533 | 2762s elapsed | 78,550 tok/s
[Eval] batch 6641 | samples 53128/123656 | mean_loss 2.3528 | 2770s elapsed | 78,551 tok/s
[Eval] batch 6661 | samples 53288/123656 | mean_loss 2.3524 | 2779s elapsed | 78,551 tok/s
[Eval] batch 6681 | samples 53448/123656 | mean_loss 2.3522 | 2787s elapsed | 78,550 tok/s
[Eval] batch 6701 | samples 53608/123656 | mean_loss 2.3518 | 2795s elapsed | 78,551 tok/s
[Eval] batch 6721 | samples 53768/123656 | mean_loss 2.3518 | 2804s elapsed | 78,550 tok/s
[Eval] batch 6741 | samples 53928/123656 | mean_loss 2.3529 | 2812s elapsed | 78,551 tok/s
[Eval] batch 6761 | samples 54088/123656 | mean_loss 2.3523 | 2820s elapsed | 78,551 tok/s
[Eval] batch 6781 | samples 54248/123656 | mean_loss 2.3522 | 2829s elapsed | 78,551 tok/s
[Eval] batch 6801 | samples 54408/123656 | mean_loss 2.3521 | 2837s elapsed | 78,552 tok/s
[Eval] batch 6821 | samples 54568/123656 | mean_loss 2.3519 | 2845s elapsed | 78,551 tok/s
[Eval] batch 6841 | samples 54728/123656 | mean_loss 2.3522 | 2854s elapsed | 78,552 tok/s
[Eval] batch 6861 | samples 54888/123656 | mean_loss 2.3522 | 2862s elapsed | 78,551 tok/s
[Eval] batch 6881 | samples 55048/123656 | mean_loss 2.3515 | 2870s elapsed | 78,550 tok/s
[Eval] batch 6901 | samples 55208/123656 | mean_loss 2.3521 | 2879s elapsed | 78,550 tok/s
[Eval] batch 6921 | samples 55368/123656 | mean_loss 2.3521 | 2887s elapsed | 78,551 tok/s
[Eval] batch 6941 | samples 55528/123656 | mean_loss 2.3516 | 2895s elapsed | 78,550 tok/s
[Eval] batch 6961 | samples 55688/123656 | mean_loss 2.3515 | 2904s elapsed | 78,551 tok/s
[Eval] batch 6981 | samples 55848/123656 | mean_loss 2.3511 | 2912s elapsed | 78,551 tok/s
[Eval] batch 7001 | samples 56008/123656 | mean_loss 2.3506 | 2920s elapsed | 78,552 tok/s
[Eval] batch 7021 | samples 56168/123656 | mean_loss 2.3499 | 2929s elapsed | 78,551 tok/s
[Eval] batch 7041 | samples 56328/123656 | mean_loss 2.3497 | 2937s elapsed | 78,550 tok/s
[Eval] batch 7061 | samples 56488/123656 | mean_loss 2.3491 | 2946s elapsed | 78,551 tok/s
[Eval] batch 7081 | samples 56648/123656 | mean_loss 2.3493 | 2954s elapsed | 78,550 tok/s
[Eval] batch 7101 | samples 56808/123656 | mean_loss 2.3493 | 2962s elapsed | 78,552 tok/s
[Eval] batch 7121 | samples 56968/123656 | mean_loss 2.3491 | 2971s elapsed | 78,551 tok/s
[Eval] batch 7141 | samples 57128/123656 | mean_loss 2.3489 | 2979s elapsed | 78,552 tok/s
[Eval] batch 7161 | samples 57288/123656 | mean_loss 2.3481 | 2987s elapsed | 78,552 tok/s
[Eval] batch 7181 | samples 57448/123656 | mean_loss 2.3473 | 2996s elapsed | 78,551 tok/s
[Eval] batch 7201 | samples 57608/123656 | mean_loss 2.3465 | 3004s elapsed | 78,552 tok/s
[Eval] batch 7221 | samples 57768/123656 | mean_loss 2.3469 | 3012s elapsed | 78,552 tok/s
[Eval] batch 7241 | samples 57928/123656 | mean_loss 2.3468 | 3021s elapsed | 78,553 tok/s
[Eval] batch 7261 | samples 58088/123656 | mean_loss 2.3461 | 3029s elapsed | 78,553 tok/s
[Eval] batch 7281 | samples 58248/123656 | mean_loss 2.3455 | 3037s elapsed | 78,553 tok/s
[Eval] batch 7301 | samples 58408/123656 | mean_loss 2.3455 | 3046s elapsed | 78,553 tok/s
[Eval] batch 7321 | samples 58568/123656 | mean_loss 2.3447 | 3054s elapsed | 78,553 tok/s
[Eval] batch 7341 | samples 58728/123656 | mean_loss 2.3439 | 3062s elapsed | 78,554 tok/s
[Eval] batch 7361 | samples 58888/123656 | mean_loss 2.3435 | 3071s elapsed | 78,554 tok/s
[Eval] batch 7381 | samples 59048/123656 | mean_loss 2.3432 | 3079s elapsed | 78,554 tok/s
[Eval] batch 7401 | samples 59208/123656 | mean_loss 2.3432 | 3087s elapsed | 78,554 tok/s
[Eval] batch 7421 | samples 59368/123656 | mean_loss 2.3432 | 3096s elapsed | 78,554 tok/s
[Eval] batch 7441 | samples 59528/123656 | mean_loss 2.3433 | 3104s elapsed | 78,554 tok/s
[Eval] batch 7461 | samples 59688/123656 | mean_loss 2.3430 | 3112s elapsed | 78,554 tok/s
[Eval] batch 7481 | samples 59848/123656 | mean_loss 2.3428 | 3121s elapsed | 78,554 tok/s
[Eval] batch 7501 | samples 60008/123656 | mean_loss 2.3428 | 3129s elapsed | 78,554 tok/s
[Eval] batch 7521 | samples 60168/123656 | mean_loss 2.3426 | 3137s elapsed | 78,555 tok/s
[Eval] batch 7541 | samples 60328/123656 | mean_loss 2.3429 | 3146s elapsed | 78,554 tok/s
[Eval] batch 7561 | samples 60488/123656 | mean_loss 2.3430 | 3154s elapsed | 78,555 tok/s
[Eval] batch 7581 | samples 60648/123656 | mean_loss 2.3426 | 3162s elapsed | 78,555 tok/s
[Eval] batch 7601 | samples 60808/123656 | mean_loss 2.3429 | 3171s elapsed | 78,555 tok/s
[Eval] batch 7621 | samples 60968/123656 | mean_loss 2.3427 | 3179s elapsed | 78,556 tok/s
[Eval] batch 7641 | samples 61128/123656 | mean_loss 2.3422 | 3187s elapsed | 78,555 tok/s
[Eval] batch 7661 | samples 61288/123656 | mean_loss 2.3421 | 3196s elapsed | 78,556 tok/s
[Eval] batch 7681 | samples 61448/123656 | mean_loss 2.3418 | 3204s elapsed | 78,555 tok/s
[Eval] batch 7701 | samples 61608/123656 | mean_loss 2.3416 | 3212s elapsed | 78,554 tok/s
[Eval] batch 7721 | samples 61768/123656 | mean_loss 2.3412 | 3221s elapsed | 78,554 tok/s
[Eval] batch 7741 | samples 61928/123656 | mean_loss 2.3409 | 3229s elapsed | 78,554 tok/s
[Eval] batch 7761 | samples 62088/123656 | mean_loss 2.3410 | 3237s elapsed | 78,553 tok/s
[Eval] batch 7781 | samples 62248/123656 | mean_loss 2.3406 | 3246s elapsed | 78,553 tok/s
[Eval] batch 7801 | samples 62408/123656 | mean_loss 2.3404 | 3254s elapsed | 78,554 tok/s
[Eval] batch 7821 | samples 62568/123656 | mean_loss 2.3405 | 3262s elapsed | 78,553 tok/s
[Eval] batch 7841 | samples 62728/123656 | mean_loss 2.3404 | 3271s elapsed | 78,554 tok/s
[Eval] batch 7861 | samples 62888/123656 | mean_loss 2.3403 | 3279s elapsed | 78,553 tok/s
[Eval] batch 7881 | samples 63048/123656 | mean_loss 2.3401 | 3288s elapsed | 78,553 tok/s
[Eval] batch 7901 | samples 63208/123656 | mean_loss 2.3404 | 3296s elapsed | 78,554 tok/s
[Eval] batch 7921 | samples 63368/123656 | mean_loss 2.3406 | 3304s elapsed | 78,553 tok/s
[Eval] batch 7941 | samples 63528/123656 | mean_loss 2.3405 | 3313s elapsed | 78,554 tok/s
[Eval] batch 7961 | samples 63688/123656 | mean_loss 2.3403 | 3321s elapsed | 78,554 tok/s
[Eval] batch 7981 | samples 63848/123656 | mean_loss 2.3401 | 3329s elapsed | 78,553 tok/s
[Eval] batch 8001 | samples 64008/123656 | mean_loss 2.3395 | 3338s elapsed | 78,554 tok/s
[Eval] batch 8021 | samples 64168/123656 | mean_loss 2.3389 | 3346s elapsed | 78,553 tok/s
[Eval] batch 8041 | samples 64328/123656 | mean_loss 2.3391 | 3354s elapsed | 78,552 tok/s
[Eval] batch 8061 | samples 64488/123656 | mean_loss 2.3388 | 3363s elapsed | 78,552 tok/s
[Eval] batch 8081 | samples 64648/123656 | mean_loss 2.3386 | 3371s elapsed | 78,552 tok/s
[Eval] batch 8101 | samples 64808/123656 | mean_loss 2.3376 | 3379s elapsed | 78,551 tok/s
[Eval] batch 8121 | samples 64968/123656 | mean_loss 2.3373 | 3388s elapsed | 78,552 tok/s
[Eval] batch 8141 | samples 65128/123656 | mean_loss 2.3373 | 3396s elapsed | 78,552 tok/s
[Eval] batch 8161 | samples 65288/123656 | mean_loss 2.3371 | 3404s elapsed | 78,553 tok/s
[Eval] batch 8181 | samples 65448/123656 | mean_loss 2.3370 | 3413s elapsed | 78,553 tok/s
[Eval] batch 8201 | samples 65608/123656 | mean_loss 2.3368 | 3421s elapsed | 78,553 tok/s
[Eval] batch 8221 | samples 65768/123656 | mean_loss 2.3371 | 3429s elapsed | 78,553 tok/s
[Eval] batch 8241 | samples 65928/123656 | mean_loss 2.3372 | 3438s elapsed | 78,554 tok/s
[Eval] batch 8261 | samples 66088/123656 | mean_loss 2.3369 | 3446s elapsed | 78,554 tok/s
[Eval] batch 8281 | samples 66248/123656 | mean_loss 2.3364 | 3454s elapsed | 78,554 tok/s
[Eval] batch 8301 | samples 66408/123656 | mean_loss 2.3367 | 3463s elapsed | 78,554 tok/s
[Eval] batch 8321 | samples 66568/123656 | mean_loss 2.3367 | 3471s elapsed | 78,555 tok/s
[Eval] batch 8341 | samples 66728/123656 | mean_loss 2.3364 | 3479s elapsed | 78,555 tok/s
[Eval] batch 8361 | samples 66888/123656 | mean_loss 2.3363 | 3488s elapsed | 78,556 tok/s
[Eval] batch 8381 | samples 67048/123656 | mean_loss 2.3362 | 3496s elapsed | 78,556 tok/s
[Eval] batch 8401 | samples 67208/123656 | mean_loss 2.3362 | 3504s elapsed | 78,555 tok/s
[Eval] batch 8421 | samples 67368/123656 | mean_loss 2.3361 | 3513s elapsed | 78,555 tok/s
[Eval] batch 8441 | samples 67528/123656 | mean_loss 2.3360 | 3521s elapsed | 78,556 tok/s
[Eval] batch 8461 | samples 67688/123656 | mean_loss 2.3359 | 3529s elapsed | 78,556 tok/s
[Eval] batch 8481 | samples 67848/123656 | mean_loss 2.3357 | 3538s elapsed | 78,556 tok/s
[Eval] batch 8501 | samples 68008/123656 | mean_loss 2.3353 | 3546s elapsed | 78,556 tok/s
[Eval] batch 8521 | samples 68168/123656 | mean_loss 2.3352 | 3554s elapsed | 78,557 tok/s
[Eval] batch 8541 | samples 68328/123656 | mean_loss 2.3352 | 3563s elapsed | 78,556 tok/s
[Eval] batch 8561 | samples 68488/123656 | mean_loss 2.3351 | 3571s elapsed | 78,555 tok/s
[Eval] batch 8581 | samples 68648/123656 | mean_loss 2.3351 | 3579s elapsed | 78,555 tok/s
[Eval] batch 8601 | samples 68808/123656 | mean_loss 2.3349 | 3588s elapsed | 78,555 tok/s
[Eval] batch 8621 | samples 68968/123656 | mean_loss 2.3349 | 3596s elapsed | 78,554 tok/s
[Eval] batch 8641 | samples 69128/123656 | mean_loss 2.3370 | 3604s elapsed | 78,554 tok/s
[Eval] batch 8661 | samples 69288/123656 | mean_loss 2.3375 | 3613s elapsed | 78,555 tok/s
[Eval] batch 8681 | samples 69448/123656 | mean_loss 2.3372 | 3621s elapsed | 78,554 tok/s
[Eval] batch 8701 | samples 69608/123656 | mean_loss 2.3371 | 3629s elapsed | 78,555 tok/s
[Eval] batch 8721 | samples 69768/123656 | mean_loss 2.3369 | 3638s elapsed | 78,554 tok/s
[Eval] batch 8741 | samples 69928/123656 | mean_loss 2.3368 | 3646s elapsed | 78,553 tok/s
[Eval] batch 8761 | samples 70088/123656 | mean_loss 2.3367 | 3655s elapsed | 78,553 tok/s
[Eval] batch 8781 | samples 70248/123656 | mean_loss 2.3361 | 3663s elapsed | 78,554 tok/s
[Eval] batch 8801 | samples 70408/123656 | mean_loss 2.3361 | 3671s elapsed | 78,554 tok/s
[Eval] batch 8821 | samples 70568/123656 | mean_loss 2.3353 | 3680s elapsed | 78,554 tok/s
[Eval] batch 8841 | samples 70728/123656 | mean_loss 2.3351 | 3688s elapsed | 78,554 tok/s
[Eval] batch 8861 | samples 70888/123656 | mean_loss 2.3350 | 3696s elapsed | 78,555 tok/s
[Eval] batch 8881 | samples 71048/123656 | mean_loss 2.3349 | 3705s elapsed | 78,555 tok/s
[Eval] batch 8901 | samples 71208/123656 | mean_loss 2.3345 | 3713s elapsed | 78,556 tok/s
[Eval] batch 8921 | samples 71368/123656 | mean_loss 2.3345 | 3721s elapsed | 78,556 tok/s
[Eval] batch 8941 | samples 71528/123656 | mean_loss 2.3354 | 3730s elapsed | 78,556 tok/s
[Eval] batch 8961 | samples 71688/123656 | mean_loss 2.3352 | 3738s elapsed | 78,555 tok/s
[Eval] batch 8981 | samples 71848/123656 | mean_loss 2.3350 | 3746s elapsed | 78,556 tok/s
[Eval] batch 9001 | samples 72008/123656 | mean_loss 2.3348 | 3755s elapsed | 78,556 tok/s
[Eval] batch 9021 | samples 72168/123656 | mean_loss 2.3344 | 3763s elapsed | 78,557 tok/s
[Eval] batch 9041 | samples 72328/123656 | mean_loss 2.3348 | 3771s elapsed | 78,556 tok/s
[Eval] batch 9061 | samples 72488/123656 | mean_loss 2.3347 | 3780s elapsed | 78,557 tok/s
[Eval] batch 9081 | samples 72648/123656 | mean_loss 2.3344 | 3788s elapsed | 78,557 tok/s
[Eval] batch 9101 | samples 72808/123656 | mean_loss 2.3342 | 3796s elapsed | 78,557 tok/s
[Eval] batch 9121 | samples 72968/123656 | mean_loss 2.3339 | 3805s elapsed | 78,558 tok/s
[Eval] batch 9141 | samples 73128/123656 | mean_loss 2.3342 | 3813s elapsed | 78,557 tok/s
[Eval] batch 9161 | samples 73288/123656 | mean_loss 2.3338 | 3821s elapsed | 78,558 tok/s
[Eval] batch 9181 | samples 73448/123656 | mean_loss 2.3337 | 3830s elapsed | 78,557 tok/s
[Eval] batch 9201 | samples 73608/123656 | mean_loss 2.3335 | 3838s elapsed | 78,556 tok/s
[Eval] batch 9221 | samples 73768/123656 | mean_loss 2.3335 | 3846s elapsed | 78,556 tok/s
[Eval] batch 9241 | samples 73928/123656 | mean_loss 2.3336 | 3855s elapsed | 78,557 tok/s
[Eval] batch 9261 | samples 74088/123656 | mean_loss 2.3336 | 3863s elapsed | 78,555 tok/s
[Eval] batch 9281 | samples 74248/123656 | mean_loss 2.3334 | 3871s elapsed | 78,555 tok/s
[Eval] batch 9301 | samples 74408/123656 | mean_loss 2.3333 | 3880s elapsed | 78,556 tok/s
[Eval] batch 9321 | samples 74568/123656 | mean_loss 2.3332 | 3888s elapsed | 78,555 tok/s
[Eval] batch 9341 | samples 74728/123656 | mean_loss 2.3331 | 3896s elapsed | 78,555 tok/s
[Eval] batch 9361 | samples 74888/123656 | mean_loss 2.3325 | 3905s elapsed | 78,556 tok/s
[Eval] batch 9381 | samples 75048/123656 | mean_loss 2.3323 | 3913s elapsed | 78,556 tok/s
[Eval] batch 9401 | samples 75208/123656 | mean_loss 2.3321 | 3921s elapsed | 78,556 tok/s
[Eval] batch 9421 | samples 75368/123656 | mean_loss 2.3319 | 3930s elapsed | 78,556 tok/s
[Eval] batch 9441 | samples 75528/123656 | mean_loss 2.3320 | 3938s elapsed | 78,557 tok/s
[Eval] batch 9461 | samples 75688/123656 | mean_loss 2.3317 | 3946s elapsed | 78,556 tok/s
[Eval] batch 9481 | samples 75848/123656 | mean_loss 2.3315 | 3955s elapsed | 78,555 tok/s
[Eval] batch 9501 | samples 76008/123656 | mean_loss 2.3315 | 3963s elapsed | 78,556 tok/s
[Eval] batch 9521 | samples 76168/123656 | mean_loss 2.3310 | 3972s elapsed | 78,555 tok/s
[Eval] batch 9541 | samples 76328/123656 | mean_loss 2.3327 | 3980s elapsed | 78,555 tok/s
[Eval] batch 9561 | samples 76488/123656 | mean_loss 2.3328 | 3988s elapsed | 78,556 tok/s
[Eval] batch 9581 | samples 76648/123656 | mean_loss 2.3325 | 3997s elapsed | 78,555 tok/s
[Eval] batch 9601 | samples 76808/123656 | mean_loss 2.3328 | 4005s elapsed | 78,556 tok/s
[Eval] batch 9621 | samples 76968/123656 | mean_loss 2.3323 | 4013s elapsed | 78,556 tok/s
[Eval] batch 9641 | samples 77128/123656 | mean_loss 2.3324 | 4022s elapsed | 78,556 tok/s
[Eval] batch 9661 | samples 77288/123656 | mean_loss 2.3325 | 4030s elapsed | 78,556 tok/s
[Eval] batch 9681 | samples 77448/123656 | mean_loss 2.3325 | 4038s elapsed | 78,555 tok/s
[Eval] batch 9701 | samples 77608/123656 | mean_loss 2.3324 | 4047s elapsed | 78,555 tok/s
[Eval] batch 9721 | samples 77768/123656 | mean_loss 2.3322 | 4055s elapsed | 78,555 tok/s
[Eval] batch 9741 | samples 77928/123656 | mean_loss 2.3318 | 4063s elapsed | 78,554 tok/s
[Eval] batch 9761 | samples 78088/123656 | mean_loss 2.3316 | 4072s elapsed | 78,555 tok/s
[Eval] batch 9781 | samples 78248/123656 | mean_loss 2.3313 | 4080s elapsed | 78,555 tok/s
[Eval] batch 9801 | samples 78408/123656 | mean_loss 2.3311 | 4088s elapsed | 78,556 tok/s
[Eval] batch 9821 | samples 78568/123656 | mean_loss 2.3311 | 4097s elapsed | 78,556 tok/s
[Eval] batch 9841 | samples 78728/123656 | mean_loss 2.3312 | 4105s elapsed | 78,557 tok/s
[Eval] batch 9861 | samples 78888/123656 | mean_loss 2.3309 | 4113s elapsed | 78,556 tok/s
[Eval] batch 9881 | samples 79048/123656 | mean_loss 2.3306 | 4122s elapsed | 78,557 tok/s
[Eval] batch 9901 | samples 79208/123656 | mean_loss 2.3303 | 4130s elapsed | 78,557 tok/s
[Eval] batch 9921 | samples 79368/123656 | mean_loss 2.3302 | 4138s elapsed | 78,556 tok/s
[Eval] batch 9941 | samples 79528/123656 | mean_loss 2.3301 | 4147s elapsed | 78,556 tok/s
[Eval] batch 9961 | samples 79688/123656 | mean_loss 2.3298 | 4155s elapsed | 78,556 tok/s
[Eval] batch 9981 | samples 79848/123656 | mean_loss 2.3300 | 4163s elapsed | 78,555 tok/s
[Eval] batch 10001 | samples 80008/123656 | mean_loss 2.3294 | 4172s elapsed | 78,556 tok/s
[Eval] batch 10021 | samples 80168/123656 | mean_loss 2.3293 | 4180s elapsed | 78,556 tok/s
[Eval] batch 10041 | samples 80328/123656 | mean_loss 2.3293 | 4188s elapsed | 78,555 tok/s
[Eval] batch 10061 | samples 80488/123656 | mean_loss 2.3293 | 4197s elapsed | 78,556 tok/s
[Eval] batch 10081 | samples 80648/123656 | mean_loss 2.3293 | 4205s elapsed | 78,555 tok/s
[Eval] batch 10101 | samples 80808/123656 | mean_loss 2.3292 | 4213s elapsed | 78,556 tok/s
[Eval] batch 10121 | samples 80968/123656 | mean_loss 2.3290 | 4222s elapsed | 78,556 tok/s
[Eval] batch 10141 | samples 81128/123656 | mean_loss 2.3287 | 4230s elapsed | 78,556 tok/s
[Eval] batch 10161 | samples 81288/123656 | mean_loss 2.3288 | 4238s elapsed | 78,556 tok/s
[Eval] batch 10181 | samples 81448/123656 | mean_loss 2.3286 | 4247s elapsed | 78,555 tok/s
[Eval] batch 10201 | samples 81608/123656 | mean_loss 2.3287 | 4255s elapsed | 78,555 tok/s
[Eval] batch 10221 | samples 81768/123656 | mean_loss 2.3293 | 4263s elapsed | 78,556 tok/s
[Eval] batch 10241 | samples 81928/123656 | mean_loss 2.3298 | 4272s elapsed | 78,555 tok/s
[Eval] batch 10261 | samples 82088/123656 | mean_loss 2.3296 | 4280s elapsed | 78,555 tok/s
[Eval] batch 10281 | samples 82248/123656 | mean_loss 2.3293 | 4289s elapsed | 78,555 tok/s
[Eval] batch 10301 | samples 82408/123656 | mean_loss 2.3289 | 4297s elapsed | 78,556 tok/s
[Eval] batch 10321 | samples 82568/123656 | mean_loss 2.3290 | 4305s elapsed | 78,556 tok/s
[Eval] batch 10341 | samples 82728/123656 | mean_loss 2.3289 | 4314s elapsed | 78,555 tok/s
[Eval] batch 10361 | samples 82888/123656 | mean_loss 2.3288 | 4322s elapsed | 78,556 tok/s
[Eval] batch 10381 | samples 83048/123656 | mean_loss 2.3288 | 4330s elapsed | 78,556 tok/s
[Eval] batch 10401 | samples 83208/123656 | mean_loss 2.3286 | 4339s elapsed | 78,556 tok/s
[Eval] batch 10421 | samples 83368/123656 | mean_loss 2.3283 | 4347s elapsed | 78,556 tok/s
[Eval] batch 10441 | samples 83528/123656 | mean_loss 2.3282 | 4355s elapsed | 78,556 tok/s
[Eval] batch 10461 | samples 83688/123656 | mean_loss 2.3280 | 4364s elapsed | 78,557 tok/s
[Eval] batch 10481 | samples 83848/123656 | mean_loss 2.3276 | 4372s elapsed | 78,557 tok/s
[Eval] batch 10501 | samples 84008/123656 | mean_loss 2.3274 | 4380s elapsed | 78,557 tok/s
[Eval] batch 10521 | samples 84168/123656 | mean_loss 2.3275 | 4389s elapsed | 78,558 tok/s
[Eval] batch 10541 | samples 84328/123656 | mean_loss 2.3272 | 4397s elapsed | 78,558 tok/s
[Eval] batch 10561 | samples 84488/123656 | mean_loss 2.3269 | 4405s elapsed | 78,558 tok/s
[Eval] batch 10581 | samples 84648/123656 | mean_loss 2.3269 | 4414s elapsed | 78,558 tok/s
[Eval] batch 10601 | samples 84808/123656 | mean_loss 2.3267 | 4422s elapsed | 78,559 tok/s
[Eval] batch 10621 | samples 84968/123656 | mean_loss 2.3262 | 4430s elapsed | 78,559 tok/s
[Eval] batch 10641 | samples 85128/123656 | mean_loss 2.3257 | 4438s elapsed | 78,559 tok/s
[Eval] batch 10661 | samples 85288/123656 | mean_loss 2.3256 | 4447s elapsed | 78,560 tok/s
[Eval] batch 10681 | samples 85448/123656 | mean_loss 2.3256 | 4455s elapsed | 78,560 tok/s
[Eval] batch 10701 | samples 85608/123656 | mean_loss 2.3254 | 4463s elapsed | 78,560 tok/s
[Eval] batch 10721 | samples 85768/123656 | mean_loss 2.3254 | 4472s elapsed | 78,559 tok/s
[Eval] batch 10741 | samples 85928/123656 | mean_loss 2.3253 | 4480s elapsed | 78,560 tok/s
[Eval] batch 10761 | samples 86088/123656 | mean_loss 2.3252 | 4489s elapsed | 78,560 tok/s
[Eval] batch 10781 | samples 86248/123656 | mean_loss 2.3248 | 4497s elapsed | 78,559 tok/s
[Eval] batch 10801 | samples 86408/123656 | mean_loss 2.3243 | 4505s elapsed | 78,558 tok/s
[Eval] batch 10821 | samples 86568/123656 | mean_loss 2.3238 | 4514s elapsed | 78,559 tok/s
[Eval] batch 10841 | samples 86728/123656 | mean_loss 2.3232 | 4522s elapsed | 78,559 tok/s
[Eval] batch 10861 | samples 86888/123656 | mean_loss 2.3227 | 4530s elapsed | 78,559 tok/s
[Eval] batch 10881 | samples 87048/123656 | mean_loss 2.3220 | 4539s elapsed | 78,559 tok/s
[Eval] batch 10901 | samples 87208/123656 | mean_loss 2.3214 | 4547s elapsed | 78,560 tok/s
[Eval] batch 10921 | samples 87368/123656 | mean_loss 2.3213 | 4555s elapsed | 78,559 tok/s
[Eval] batch 10941 | samples 87528/123656 | mean_loss 2.3209 | 4564s elapsed | 78,559 tok/s
[Eval] batch 10961 | samples 87688/123656 | mean_loss 2.3202 | 4572s elapsed | 78,560 tok/s
[Eval] batch 10981 | samples 87848/123656 | mean_loss 2.3197 | 4580s elapsed | 78,560 tok/s
[Eval] batch 11001 | samples 88008/123656 | mean_loss 2.3193 | 4589s elapsed | 78,560 tok/s
[Eval] batch 11021 | samples 88168/123656 | mean_loss 2.3188 | 4597s elapsed | 78,560 tok/s
[Eval] batch 11041 | samples 88328/123656 | mean_loss 2.3183 | 4605s elapsed | 78,560 tok/s
[Eval] batch 11061 | samples 88488/123656 | mean_loss 2.3173 | 4614s elapsed | 78,560 tok/s
[Eval] batch 11081 | samples 88648/123656 | mean_loss 2.3171 | 4622s elapsed | 78,560 tok/s
[Eval] batch 11101 | samples 88808/123656 | mean_loss 2.3166 | 4630s elapsed | 78,561 tok/s
[Eval] batch 11121 | samples 88968/123656 | mean_loss 2.3161 | 4639s elapsed | 78,561 tok/s
[Eval] batch 11141 | samples 89128/123656 | mean_loss 2.3158 | 4647s elapsed | 78,561 tok/s
[Eval] batch 11161 | samples 89288/123656 | mean_loss 2.3153 | 4655s elapsed | 78,560 tok/s
[Eval] batch 11181 | samples 89448/123656 | mean_loss 2.3147 | 4664s elapsed | 78,560 tok/s
[Eval] batch 11201 | samples 89608/123656 | mean_loss 2.3142 | 4672s elapsed | 78,561 tok/s
[Eval] batch 11221 | samples 89768/123656 | mean_loss 2.3136 | 4680s elapsed | 78,560 tok/s
[Eval] batch 11241 | samples 89928/123656 | mean_loss 2.3133 | 4689s elapsed | 78,560 tok/s
[Eval] batch 11261 | samples 90088/123656 | mean_loss 2.3128 | 4697s elapsed | 78,560 tok/s
[Eval] batch 11281 | samples 90248/123656 | mean_loss 2.3121 | 4705s elapsed | 78,560 tok/s
[Eval] batch 11301 | samples 90408/123656 | mean_loss 2.3119 | 4714s elapsed | 78,561 tok/s
[Eval] batch 11321 | samples 90568/123656 | mean_loss 2.3114 | 4722s elapsed | 78,561 tok/s
[Eval] batch 11341 | samples 90728/123656 | mean_loss 2.3119 | 4730s elapsed | 78,561 tok/s
[Eval] batch 11361 | samples 90888/123656 | mean_loss 2.3120 | 4739s elapsed | 78,560 tok/s
[Eval] batch 11381 | samples 91048/123656 | mean_loss 2.3113 | 4747s elapsed | 78,560 tok/s
[Eval] batch 11401 | samples 91208/123656 | mean_loss 2.3106 | 4755s elapsed | 78,561 tok/s
[Eval] batch 11421 | samples 91368/123656 | mean_loss 2.3100 | 4764s elapsed | 78,561 tok/s
[Eval] batch 11441 | samples 91528/123656 | mean_loss 2.3095 | 4772s elapsed | 78,561 tok/s
[Eval] batch 11461 | samples 91688/123656 | mean_loss 2.3087 | 4780s elapsed | 78,561 tok/s
[Eval] batch 11481 | samples 91848/123656 | mean_loss 2.3082 | 4789s elapsed | 78,561 tok/s
[Eval] batch 11501 | samples 92008/123656 | mean_loss 2.3075 | 4797s elapsed | 78,561 tok/s
[Eval] batch 11521 | samples 92168/123656 | mean_loss 2.3070 | 4805s elapsed | 78,561 tok/s
[Eval] batch 11541 | samples 92328/123656 | mean_loss 2.3065 | 4814s elapsed | 78,561 tok/s
[Eval] batch 11561 | samples 92488/123656 | mean_loss 2.3058 | 4822s elapsed | 78,562 tok/s
[Eval] batch 11581 | samples 92648/123656 | mean_loss 2.3052 | 4830s elapsed | 78,562 tok/s
[Eval] batch 11601 | samples 92808/123656 | mean_loss 2.3046 | 4839s elapsed | 78,562 tok/s
[Eval] batch 11621 | samples 92968/123656 | mean_loss 2.3039 | 4847s elapsed | 78,562 tok/s
[Eval] batch 11641 | samples 93128/123656 | mean_loss 2.3034 | 4855s elapsed | 78,563 tok/s
[Eval] batch 11661 | samples 93288/123656 | mean_loss 2.3027 | 4864s elapsed | 78,563 tok/s
[Eval] batch 11681 | samples 93448/123656 | mean_loss 2.3021 | 4872s elapsed | 78,562 tok/s
[Eval] batch 11701 | samples 93608/123656 | mean_loss 2.3014 | 4880s elapsed | 78,562 tok/s
[Eval] batch 11721 | samples 93768/123656 | mean_loss 2.3011 | 4889s elapsed | 78,562 tok/s
[Eval] batch 11741 | samples 93928/123656 | mean_loss 2.3008 | 4897s elapsed | 78,562 tok/s
[Eval] batch 11761 | samples 94088/123656 | mean_loss 2.3008 | 4906s elapsed | 78,561 tok/s
[Eval] batch 11781 | samples 94248/123656 | mean_loss 2.3000 | 4914s elapsed | 78,562 tok/s
[Eval] batch 11801 | samples 94408/123656 | mean_loss 2.2997 | 4922s elapsed | 78,561 tok/s
[Eval] batch 11821 | samples 94568/123656 | mean_loss 2.2995 | 4931s elapsed | 78,561 tok/s
[Eval] batch 11841 | samples 94728/123656 | mean_loss 2.2987 | 4939s elapsed | 78,561 tok/s
[Eval] batch 11861 | samples 94888/123656 | mean_loss 2.2979 | 4947s elapsed | 78,561 tok/s
[Eval] batch 11881 | samples 95048/123656 | mean_loss 2.2975 | 4956s elapsed | 78,562 tok/s
[Eval] batch 11901 | samples 95208/123656 | mean_loss 2.2969 | 4964s elapsed | 78,561 tok/s
[Eval] batch 11921 | samples 95368/123656 | mean_loss 2.2964 | 4972s elapsed | 78,562 tok/s
[Eval] batch 11941 | samples 95528/123656 | mean_loss 2.2960 | 4981s elapsed | 78,562 tok/s
[Eval] batch 11961 | samples 95688/123656 | mean_loss 2.2955 | 4989s elapsed | 78,561 tok/s
[Eval] batch 11981 | samples 95848/123656 | mean_loss 2.2949 | 4997s elapsed | 78,561 tok/s
[Eval] batch 12001 | samples 96008/123656 | mean_loss 2.2943 | 5006s elapsed | 78,561 tok/s
[Eval] batch 12021 | samples 96168/123656 | mean_loss 2.2926 | 5014s elapsed | 78,561 tok/s
[Eval] batch 12041 | samples 96328/123656 | mean_loss 2.2915 | 5022s elapsed | 78,562 tok/s
[Eval] batch 12061 | samples 96488/123656 | mean_loss 2.2918 | 5031s elapsed | 78,562 tok/s
[Eval] batch 12081 | samples 96648/123656 | mean_loss 2.2916 | 5039s elapsed | 78,562 tok/s
[Eval] batch 12101 | samples 96808/123656 | mean_loss 2.2934 | 5047s elapsed | 78,561 tok/s
[Eval] batch 12121 | samples 96968/123656 | mean_loss 2.2933 | 5056s elapsed | 78,562 tok/s
[Eval] batch 12141 | samples 97128/123656 | mean_loss 2.2942 | 5064s elapsed | 78,562 tok/s
[Eval] batch 12161 | samples 97288/123656 | mean_loss 2.2951 | 5072s elapsed | 78,562 tok/s
[Eval] batch 12181 | samples 97448/123656 | mean_loss 2.2950 | 5081s elapsed | 78,562 tok/s
[Eval] batch 12201 | samples 97608/123656 | mean_loss 2.2960 | 5089s elapsed | 78,563 tok/s
[Eval] batch 12221 | samples 97768/123656 | mean_loss 2.2971 | 5097s elapsed | 78,563 tok/s
[Eval] batch 12241 | samples 97928/123656 | mean_loss 2.2972 | 5106s elapsed | 78,563 tok/s
[Eval] batch 12261 | samples 98088/123656 | mean_loss 2.2966 | 5114s elapsed | 78,563 tok/s
[Eval] batch 12281 | samples 98248/123656 | mean_loss 2.2960 | 5122s elapsed | 78,563 tok/s
[Eval] batch 12301 | samples 98408/123656 | mean_loss 2.2975 | 5131s elapsed | 78,564 tok/s
[Eval] batch 12321 | samples 98568/123656 | mean_loss 2.2974 | 5139s elapsed | 78,564 tok/s
[Eval] batch 12341 | samples 98728/123656 | mean_loss 2.2983 | 5147s elapsed | 78,564 tok/s
[Eval] batch 12361 | samples 98888/123656 | mean_loss 2.2978 | 5156s elapsed | 78,564 tok/s
[Eval] batch 12381 | samples 99048/123656 | mean_loss 2.2980 | 5164s elapsed | 78,564 tok/s
[Eval] batch 12401 | samples 99208/123656 | mean_loss 2.2981 | 5172s elapsed | 78,565 tok/s
[Eval] batch 12421 | samples 99368/123656 | mean_loss 2.2987 | 5181s elapsed | 78,565 tok/s
[Eval] batch 12441 | samples 99528/123656 | mean_loss 2.2991 | 5189s elapsed | 78,565 tok/s
[Eval] batch 12461 | samples 99688/123656 | mean_loss 2.2986 | 5197s elapsed | 78,565 tok/s
[Eval] batch 12481 | samples 99848/123656 | mean_loss 2.2982 | 5206s elapsed | 78,565 tok/s
[Eval] batch 12501 | samples 100008/123656 | mean_loss 2.2973 | 5214s elapsed | 78,565 tok/s
[Eval] batch 12521 | samples 100168/123656 | mean_loss 2.2963 | 5222s elapsed | 78,565 tok/s
[Eval] batch 12541 | samples 100328/123656 | mean_loss 2.2963 | 5231s elapsed | 78,566 tok/s
[Eval] batch 12561 | samples 100488/123656 | mean_loss 2.2960 | 5239s elapsed | 78,565 tok/s
[Eval] batch 12581 | samples 100648/123656 | mean_loss 2.2963 | 5247s elapsed | 78,565 tok/s
[Eval] batch 12601 | samples 100808/123656 | mean_loss 2.2963 | 5256s elapsed | 78,566 tok/s
[Eval] batch 12621 | samples 100968/123656 | mean_loss 2.2960 | 5264s elapsed | 78,565 tok/s
[Eval] batch 12641 | samples 101128/123656 | mean_loss 2.2962 | 5272s elapsed | 78,565 tok/s
[Eval] batch 12661 | samples 101288/123656 | mean_loss 2.2969 | 5281s elapsed | 78,566 tok/s
[Eval] batch 12681 | samples 101448/123656 | mean_loss 2.2965 | 5289s elapsed | 78,566 tok/s
[Eval] batch 12701 | samples 101608/123656 | mean_loss 2.2953 | 5297s elapsed | 78,566 tok/s
[Eval] batch 12721 | samples 101768/123656 | mean_loss 2.2951 | 5306s elapsed | 78,566 tok/s
[Eval] batch 12741 | samples 101928/123656 | mean_loss 2.2944 | 5314s elapsed | 78,566 tok/s
[Eval] batch 12761 | samples 102088/123656 | mean_loss 2.2947 | 5322s elapsed | 78,566 tok/s
[Eval] batch 12781 | samples 102248/123656 | mean_loss 2.2964 | 5331s elapsed | 78,567 tok/s
[Eval] batch 12801 | samples 102408/123656 | mean_loss 2.2965 | 5339s elapsed | 78,567 tok/s
[Eval] batch 12821 | samples 102568/123656 | mean_loss 2.2962 | 5347s elapsed | 78,567 tok/s
[Eval] batch 12841 | samples 102728/123656 | mean_loss 2.2956 | 5356s elapsed | 78,567 tok/s
[Eval] batch 12861 | samples 102888/123656 | mean_loss 2.2960 | 5364s elapsed | 78,568 tok/s
[Eval] batch 12881 | samples 103048/123656 | mean_loss 2.2969 | 5372s elapsed | 78,567 tok/s
[Eval] batch 12901 | samples 103208/123656 | mean_loss 2.2969 | 5381s elapsed | 78,568 tok/s
[Eval] batch 12921 | samples 103368/123656 | mean_loss 2.2961 | 5389s elapsed | 78,568 tok/s
[Eval] batch 12941 | samples 103528/123656 | mean_loss 2.2958 | 5397s elapsed | 78,568 tok/s
[Eval] batch 12961 | samples 103688/123656 | mean_loss 2.2952 | 5406s elapsed | 78,568 tok/s
[Eval] batch 12981 | samples 103848/123656 | mean_loss 2.2955 | 5414s elapsed | 78,568 tok/s
[Eval] batch 13001 | samples 104008/123656 | mean_loss 2.2955 | 5422s elapsed | 78,568 tok/s
[Eval] batch 13021 | samples 104168/123656 | mean_loss 2.2949 | 5431s elapsed | 78,567 tok/s
[Eval] batch 13041 | samples 104328/123656 | mean_loss 2.2958 | 5439s elapsed | 78,568 tok/s
[Eval] batch 13061 | samples 104488/123656 | mean_loss 2.2987 | 5447s elapsed | 78,568 tok/s
[Eval] batch 13081 | samples 104648/123656 | mean_loss 2.3003 | 5456s elapsed | 78,569 tok/s
[Eval] batch 13101 | samples 104808/123656 | mean_loss 2.3011 | 5464s elapsed | 78,568 tok/s
[Eval] batch 13121 | samples 104968/123656 | mean_loss 2.3007 | 5472s elapsed | 78,569 tok/s
[Eval] batch 13141 | samples 105128/123656 | mean_loss 2.3009 | 5481s elapsed | 78,569 tok/s
[Eval] batch 13161 | samples 105288/123656 | mean_loss 2.3011 | 5489s elapsed | 78,569 tok/s
[Eval] batch 13181 | samples 105448/123656 | mean_loss 2.3010 | 5497s elapsed | 78,569 tok/s
[Eval] batch 13201 | samples 105608/123656 | mean_loss 2.3005 | 5506s elapsed | 78,570 tok/s
[Eval] batch 13221 | samples 105768/123656 | mean_loss 2.3005 | 5514s elapsed | 78,570 tok/s
[Eval] batch 13241 | samples 105928/123656 | mean_loss 2.3011 | 5522s elapsed | 78,570 tok/s
[Eval] batch 13261 | samples 106088/123656 | mean_loss 2.3006 | 5531s elapsed | 78,571 tok/s
[Eval] batch 13281 | samples 106248/123656 | mean_loss 2.3005 | 5539s elapsed | 78,571 tok/s
[Eval] batch 13301 | samples 106408/123656 | mean_loss 2.3004 | 5547s elapsed | 78,571 tok/s
[Eval] batch 13321 | samples 106568/123656 | mean_loss 2.2997 | 5556s elapsed | 78,571 tok/s
[Eval] batch 13341 | samples 106728/123656 | mean_loss 2.2987 | 5564s elapsed | 78,571 tok/s
[Eval] batch 13361 | samples 106888/123656 | mean_loss 2.2997 | 5572s elapsed | 78,571 tok/s
[Eval] batch 13381 | samples 107048/123656 | mean_loss 2.2994 | 5581s elapsed | 78,571 tok/s
[Eval] batch 13401 | samples 107208/123656 | mean_loss 2.2984 | 5589s elapsed | 78,572 tok/s
[Eval] batch 13421 | samples 107368/123656 | mean_loss 2.2983 | 5597s elapsed | 78,571 tok/s
[Eval] batch 13441 | samples 107528/123656 | mean_loss 2.2996 | 5606s elapsed | 78,571 tok/s
[Eval] batch 13461 | samples 107688/123656 | mean_loss 2.3005 | 5614s elapsed | 78,572 tok/s
[Eval] batch 13481 | samples 107848/123656 | mean_loss 2.2996 | 5622s elapsed | 78,572 tok/s
[Eval] batch 13501 | samples 108008/123656 | mean_loss 2.2994 | 5631s elapsed | 78,572 tok/s
[Eval] batch 13521 | samples 108168/123656 | mean_loss 2.2993 | 5639s elapsed | 78,572 tok/s
[Eval] batch 13541 | samples 108328/123656 | mean_loss 2.3016 | 5647s elapsed | 78,572 tok/s
[Eval] batch 13561 | samples 108488/123656 | mean_loss 2.3017 | 5656s elapsed | 78,571 tok/s
[Eval] batch 13581 | samples 108648/123656 | mean_loss 2.3010 | 5664s elapsed | 78,571 tok/s
[Eval] batch 13601 | samples 108808/123656 | mean_loss 2.3012 | 5672s elapsed | 78,572 tok/s
[Eval] batch 13621 | samples 108968/123656 | mean_loss 2.3002 | 5681s elapsed | 78,572 tok/s
[Eval] batch 13641 | samples 109128/123656 | mean_loss 2.3008 | 5689s elapsed | 78,572 tok/s
[Eval] batch 13661 | samples 109288/123656 | mean_loss 2.3024 | 5697s elapsed | 78,572 tok/s
[Eval] batch 13681 | samples 109448/123656 | mean_loss 2.3029 | 5706s elapsed | 78,572 tok/s
[Eval] batch 13701 | samples 109608/123656 | mean_loss 2.3034 | 5714s elapsed | 78,572 tok/s
[Eval] batch 13721 | samples 109768/123656 | mean_loss 2.3038 | 5722s elapsed | 78,572 tok/s
[Eval] batch 13741 | samples 109928/123656 | mean_loss 2.3031 | 5731s elapsed | 78,572 tok/s
[Eval] batch 13761 | samples 110088/123656 | mean_loss 2.3036 | 5739s elapsed | 78,572 tok/s
[Eval] batch 13781 | samples 110248/123656 | mean_loss 2.3028 | 5747s elapsed | 78,572 tok/s
[Eval] batch 13801 | samples 110408/123656 | mean_loss 2.3024 | 5756s elapsed | 78,573 tok/s
[Eval] batch 13821 | samples 110568/123656 | mean_loss 2.3054 | 5764s elapsed | 78,573 tok/s
[Eval] batch 13841 | samples 110728/123656 | mean_loss 2.3049 | 5772s elapsed | 78,573 tok/s
[Eval] batch 13861 | samples 110888/123656 | mean_loss 2.3051 | 5781s elapsed | 78,573 tok/s
[Eval] batch 13881 | samples 111048/123656 | mean_loss 2.3046 | 5789s elapsed | 78,573 tok/s
[Eval] batch 13901 | samples 111208/123656 | mean_loss 2.3053 | 5797s elapsed | 78,574 tok/s
[Eval] batch 13921 | samples 111368/123656 | mean_loss 2.3055 | 5806s elapsed | 78,574 tok/s
[Eval] batch 13941 | samples 111528/123656 | mean_loss 2.3051 | 5814s elapsed | 78,574 tok/s
[Eval] batch 13961 | samples 111688/123656 | mean_loss 2.3050 | 5822s elapsed | 78,573 tok/s
[Eval] batch 13981 | samples 111848/123656 | mean_loss 2.3047 | 5831s elapsed | 78,574 tok/s
[Eval] batch 14001 | samples 112008/123656 | mean_loss 2.3048 | 5839s elapsed | 78,574 tok/s
[Eval] batch 14021 | samples 112168/123656 | mean_loss 2.3044 | 5847s elapsed | 78,574 tok/s
[Eval] batch 14041 | samples 112328/123656 | mean_loss 2.3050 | 5856s elapsed | 78,574 tok/s
[Eval] batch 14061 | samples 112488/123656 | mean_loss 2.3048 | 5864s elapsed | 78,574 tok/s
[Eval] batch 14081 | samples 112648/123656 | mean_loss 2.3049 | 5872s elapsed | 78,574 tok/s
[Eval] batch 14101 | samples 112808/123656 | mean_loss 2.3046 | 5881s elapsed | 78,575 tok/s
[Eval] batch 14121 | samples 112968/123656 | mean_loss 2.3043 | 5889s elapsed | 78,575 tok/s
[Eval] batch 14141 | samples 113128/123656 | mean_loss 2.3052 | 5897s elapsed | 78,575 tok/s
[Eval] batch 14161 | samples 113288/123656 | mean_loss 2.3048 | 5906s elapsed | 78,575 tok/s
[Eval] batch 14181 | samples 113448/123656 | mean_loss 2.3052 | 5914s elapsed | 78,575 tok/s
[Eval] batch 14201 | samples 113608/123656 | mean_loss 2.3048 | 5922s elapsed | 78,575 tok/s
[Eval] batch 14221 | samples 113768/123656 | mean_loss 2.3054 | 5931s elapsed | 78,575 tok/s
[Eval] batch 14241 | samples 113928/123656 | mean_loss 2.3048 | 5939s elapsed | 78,576 tok/s
[Eval] batch 14261 | samples 114088/123656 | mean_loss 2.3046 | 5947s elapsed | 78,575 tok/s
[Eval] batch 14281 | samples 114248/123656 | mean_loss 2.3048 | 5956s elapsed | 78,576 tok/s
[Eval] batch 14301 | samples 114408/123656 | mean_loss 2.3046 | 5964s elapsed | 78,575 tok/s
[Eval] batch 14321 | samples 114568/123656 | mean_loss 2.3051 | 5972s elapsed | 78,576 tok/s
[Eval] batch 14341 | samples 114728/123656 | mean_loss 2.3060 | 5981s elapsed | 78,576 tok/s
[Eval] batch 14361 | samples 114888/123656 | mean_loss 2.3071 | 5989s elapsed | 78,577 tok/s
[Eval] batch 14381 | samples 115048/123656 | mean_loss 2.3074 | 5997s elapsed | 78,576 tok/s
[Eval] batch 14401 | samples 115208/123656 | mean_loss 2.3079 | 6005s elapsed | 78,577 tok/s
[Eval] batch 14421 | samples 115368/123656 | mean_loss 2.3077 | 6014s elapsed | 78,576 tok/s
[Eval] batch 14441 | samples 115528/123656 | mean_loss 2.3079 | 6022s elapsed | 78,577 tok/s
[Eval] batch 14461 | samples 115688/123656 | mean_loss 2.3083 | 6031s elapsed | 78,577 tok/s
[Eval] batch 14481 | samples 115848/123656 | mean_loss 2.3081 | 6039s elapsed | 78,577 tok/s
[Eval] batch 14501 | samples 116008/123656 | mean_loss 2.3079 | 6047s elapsed | 78,577 tok/s
[Eval] batch 14521 | samples 116168/123656 | mean_loss 2.3081 | 6055s elapsed | 78,578 tok/s
[Eval] batch 14541 | samples 116328/123656 | mean_loss 2.3082 | 6064s elapsed | 78,578 tok/s
[Eval] batch 14561 | samples 116488/123656 | mean_loss 2.3081 | 6072s elapsed | 78,578 tok/s
[Eval] batch 14581 | samples 116648/123656 | mean_loss 2.3084 | 6080s elapsed | 78,578 tok/s
[Eval] batch 14601 | samples 116808/123656 | mean_loss 2.3080 | 6089s elapsed | 78,577 tok/s
[Eval] batch 14621 | samples 116968/123656 | mean_loss 2.3078 | 6097s elapsed | 78,577 tok/s
[Eval] batch 14641 | samples 117128/123656 | mean_loss 2.3084 | 6106s elapsed | 78,577 tok/s
[Eval] batch 14661 | samples 117288/123656 | mean_loss 2.3088 | 6114s elapsed | 78,578 tok/s
[Eval] batch 14681 | samples 117448/123656 | mean_loss 2.3087 | 6122s elapsed | 78,578 tok/s
[Eval] batch 14701 | samples 117608/123656 | mean_loss 2.3096 | 6131s elapsed | 78,578 tok/s
[Eval] batch 14721 | samples 117768/123656 | mean_loss 2.3091 | 6139s elapsed | 78,578 tok/s
[Eval] batch 14741 | samples 117928/123656 | mean_loss 2.3092 | 6147s elapsed | 78,578 tok/s
[Eval] batch 14761 | samples 118088/123656 | mean_loss 2.3104 | 6156s elapsed | 78,578 tok/s
[Eval] batch 14781 | samples 118248/123656 | mean_loss 2.3119 | 6164s elapsed | 78,578 tok/s
[Eval] batch 14801 | samples 118408/123656 | mean_loss 2.3121 | 6172s elapsed | 78,579 tok/s
[Eval] batch 14821 | samples 118568/123656 | mean_loss 2.3129 | 6181s elapsed | 78,578 tok/s
[Eval] batch 14841 | samples 118728/123656 | mean_loss 2.3129 | 6189s elapsed | 78,578 tok/s
[Eval] batch 14861 | samples 118888/123656 | mean_loss 2.3124 | 6197s elapsed | 78,578 tok/s
[Eval] batch 14881 | samples 119048/123656 | mean_loss 2.3123 | 6205s elapsed | 78,579 tok/s
[Eval] batch 14901 | samples 119208/123656 | mean_loss 2.3119 | 6214s elapsed | 78,579 tok/s
[Eval] batch 14921 | samples 119368/123656 | mean_loss 2.3120 | 6222s elapsed | 78,579 tok/s
[Eval] batch 14941 | samples 119528/123656 | mean_loss 2.3114 | 6231s elapsed | 78,579 tok/s
[Eval] batch 14961 | samples 119688/123656 | mean_loss 2.3113 | 6239s elapsed | 78,579 tok/s
[Eval] batch 14981 | samples 119848/123656 | mean_loss 2.3116 | 6247s elapsed | 78,579 tok/s
[Eval] batch 15001 | samples 120008/123656 | mean_loss 2.3111 | 6256s elapsed | 78,579 tok/s
[Eval] batch 15021 | samples 120168/123656 | mean_loss 2.3110 | 6264s elapsed | 78,579 tok/s
[Eval] batch 15041 | samples 120328/123656 | mean_loss 2.3113 | 6272s elapsed | 78,579 tok/s
[Eval] batch 15061 | samples 120488/123656 | mean_loss 2.3114 | 6281s elapsed | 78,579 tok/s
[Eval] batch 15081 | samples 120648/123656 | mean_loss 2.3108 | 6289s elapsed | 78,579 tok/s
[Eval] batch 15101 | samples 120808/123656 | mean_loss 2.3106 | 6297s elapsed | 78,580 tok/s
[Eval] batch 15121 | samples 120968/123656 | mean_loss 2.3109 | 6306s elapsed | 78,579 tok/s
[Eval] batch 15141 | samples 121128/123656 | mean_loss 2.3101 | 6314s elapsed | 78,580 tok/s
[Eval] batch 15161 | samples 121288/123656 | mean_loss 2.3099 | 6322s elapsed | 78,580 tok/s
[Eval] batch 15181 | samples 121448/123656 | mean_loss 2.3107 | 6331s elapsed | 78,580 tok/s
[Eval] batch 15201 | samples 121608/123656 | mean_loss 2.3110 | 6339s elapsed | 78,580 tok/s
[Eval] batch 15221 | samples 121768/123656 | mean_loss 2.3125 | 6347s elapsed | 78,579 tok/s
[Eval] batch 15241 | samples 121928/123656 | mean_loss 2.3128 | 6356s elapsed | 78,580 tok/s
[Eval] batch 15261 | samples 122088/123656 | mean_loss 2.3132 | 6364s elapsed | 78,580 tok/s
[Eval] batch 15281 | samples 122248/123656 | mean_loss 2.3137 | 6372s elapsed | 78,580 tok/s
[Eval] batch 15301 | samples 122408/123656 | mean_loss 2.3129 | 6381s elapsed | 78,580 tok/s
[Eval] batch 15321 | samples 122568/123656 | mean_loss 2.3152 | 6389s elapsed | 78,581 tok/s
[Eval] batch 15341 | samples 122728/123656 | mean_loss 2.3147 | 6397s elapsed | 78,580 tok/s
[Eval] batch 15361 | samples 122888/123656 | mean_loss 2.3145 | 6406s elapsed | 78,581 tok/s
[Eval] batch 15381 | samples 123048/123656 | mean_loss 2.3156 | 6414s elapsed | 78,581 tok/s
[Eval] batch 15401 | samples 123208/123656 | mean_loss 2.3159 | 6422s elapsed | 78,581 tok/s
[Eval] batch 15421 | samples 123368/123656 | mean_loss 2.3156 | 6431s elapsed | 78,581 tok/s
[Eval] batch 15441 | samples 123528/123656 | mean_loss 2.3159 | 6439s elapsed | 78,581 tok/s
[Eval] batch 15457 | samples 123656/123656 | mean_loss 2.3163 | 6445s elapsed | 78,581 tok/s
[Eval] DONE in 6446s
[Eval] samples=123656 tokens=506494976
[Eval] mean_eval_loss = 2.316302
[Eval] perplexity = 10.138