File size: 116,708 Bytes
d56c069 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 |
[2023-04-14 06:57:51,009] [WARNING] [runner.py:190:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only.
[2023-04-14 06:57:51,024] [INFO] [runner.py:540:main] cmd = /usr/local/bin/python -u -m deepspeed.launcher.launch --world_info=eyJsb2NhbGhvc3QiOiBbMF19 --master_addr=127.0.0.1 --master_port=29500 --enable_each_rank_log=None main.py --model_name_or_path facebook/opt-1.3b --gradient_accumulation_steps 2 --lora_dim 128 --zero_stage 0 --deepspeed --output_dir /output/DeepSpeedExamples/applications/DeepSpeed-Chat/output/actor-models/1.3b
[2023-04-14 06:57:54,173] [INFO] [launch.py:222:main] 0 NV_LIBNCCL_DEV_PACKAGE=libnccl-dev=2.13.4-1+cuda11.7
[2023-04-14 06:57:54,173] [INFO] [launch.py:222:main] 0 NCCL_VERSION=2.13.4-1
[2023-04-14 06:57:54,173] [INFO] [launch.py:222:main] 0 NV_LIBNCCL_PACKAGE_VERSION=2.13.4-1
[2023-04-14 06:57:54,173] [INFO] [launch.py:222:main] 0 NV_LIBNCCL_PACKAGE=libnccl2=2.13.4-1+cuda11.7
[2023-04-14 06:57:54,173] [INFO] [launch.py:222:main] 0 NV_LIBNCCL_DEV_PACKAGE_NAME=libnccl-dev
[2023-04-14 06:57:54,173] [INFO] [launch.py:222:main] 0 NV_LIBNCCL_PACKAGE_NAME=libnccl2
[2023-04-14 06:57:54,173] [INFO] [launch.py:222:main] 0 NV_LIBNCCL_DEV_PACKAGE_VERSION=2.13.4-1
[2023-04-14 06:57:54,173] [INFO] [launch.py:229:main] WORLD INFO DICT: {'localhost': [0]}
[2023-04-14 06:57:54,173] [INFO] [launch.py:235:main] nnodes=1, num_local_procs=1, node_rank=0
[2023-04-14 06:57:54,173] [INFO] [launch.py:246:main] global_rank_mapping=defaultdict(<class 'list'>, {'localhost': [0]})
[2023-04-14 06:57:54,173] [INFO] [launch.py:247:main] dist_world_size=1
[2023-04-14 06:57:54,173] [INFO] [launch.py:249:main] Setting CUDA_VISIBLE_DEVICES=0
[2023-04-14 06:57:58,372] [INFO] [comm.py:586:init_distributed] Initializing TorchBackend in DeepSpeed with backend nccl
Downloading (…)okenizer_config.json: 0%| | 0.00/685 [00:00<?, ?B/s]
Downloading (…)okenizer_config.json: 100%|██████████| 685/685 [00:00<00:00, 46.7kB/s]
Downloading (…)lve/main/config.json: 0%| | 0.00/653 [00:00<?, ?B/s]
Downloading (…)lve/main/config.json: 100%|██████████| 653/653 [00:00<00:00, 196kB/s]
Downloading (…)olve/main/vocab.json: 0%| | 0.00/899k [00:00<?, ?B/s]
Downloading (…)olve/main/vocab.json: 100%|██████████| 899k/899k [00:00<00:00, 1.01MB/s]
Downloading (…)olve/main/vocab.json: 100%|██████████| 899k/899k [00:00<00:00, 1.01MB/s]
Downloading (…)olve/main/merges.txt: 0%| | 0.00/456k [00:00<?, ?B/s]
Downloading (…)olve/main/merges.txt: 100%|██████████| 456k/456k [00:00<00:00, 645kB/s]
Downloading (…)olve/main/merges.txt: 100%|██████████| 456k/456k [00:00<00:00, 643kB/s]
Downloading (…)cial_tokens_map.json: 0%| | 0.00/441 [00:00<?, ?B/s]
Downloading (…)cial_tokens_map.json: 100%|██████████| 441/441 [00:00<00:00, 90.2kB/s]
Downloading pytorch_model.bin: 0%| | 0.00/2.63G [00:00<?, ?B/s]
Downloading pytorch_model.bin: 0%| | 10.5M/2.63G [00:00<01:57, 22.3MB/s]
Downloading pytorch_model.bin: 1%| | 21.0M/2.63G [00:00<01:19, 32.9MB/s]
Downloading pytorch_model.bin: 1%| | 31.5M/2.63G [00:00<01:05, 39.6MB/s]
Downloading pytorch_model.bin: 2%|▏ | 41.9M/2.63G [00:01<00:58, 44.0MB/s]
Downloading pytorch_model.bin: 2%|▏ | 52.4M/2.63G [00:01<00:55, 46.3MB/s]
Downloading pytorch_model.bin: 2%|▏ | 62.9M/2.63G [00:01<00:53, 47.7MB/s]
Downloading pytorch_model.bin: 3%|▎ | 73.4M/2.63G [00:01<00:52, 49.1MB/s]
Downloading pytorch_model.bin: 3%|▎ | 83.9M/2.63G [00:01<00:51, 49.7MB/s]
Downloading pytorch_model.bin: 4%|▎ | 94.4M/2.63G [00:02<00:50, 50.5MB/s]
Downloading pytorch_model.bin: 4%|▍ | 105M/2.63G [00:02<00:48, 52.0MB/s]
Downloading pytorch_model.bin: 4%|▍ | 115M/2.63G [00:02<00:48, 51.8MB/s]
Downloading pytorch_model.bin: 5%|▍ | 126M/2.63G [00:02<00:48, 51.8MB/s]
Downloading pytorch_model.bin: 5%|▌ | 136M/2.63G [00:02<00:47, 52.2MB/s]
Downloading pytorch_model.bin: 6%|▌ | 147M/2.63G [00:03<00:48, 51.4MB/s]
Downloading pytorch_model.bin: 6%|▌ | 157M/2.63G [00:03<00:47, 52.0MB/s]
Downloading pytorch_model.bin: 6%|▋ | 168M/2.63G [00:03<00:47, 51.6MB/s]
Downloading pytorch_model.bin: 7%|▋ | 178M/2.63G [00:03<00:47, 51.6MB/s]
Downloading pytorch_model.bin: 7%|▋ | 189M/2.63G [00:03<00:47, 51.8MB/s]
Downloading pytorch_model.bin: 8%|▊ | 199M/2.63G [00:04<00:48, 49.8MB/s]
Downloading pytorch_model.bin: 8%|▊ | 210M/2.63G [00:04<00:48, 50.4MB/s]
Downloading pytorch_model.bin: 8%|▊ | 220M/2.63G [00:04<00:47, 51.0MB/s]
Downloading pytorch_model.bin: 9%|▉ | 231M/2.63G [00:04<00:46, 51.8MB/s]
Downloading pytorch_model.bin: 9%|▉ | 241M/2.63G [00:04<00:43, 55.4MB/s]
Downloading pytorch_model.bin: 10%|▉ | 252M/2.63G [00:05<00:43, 54.5MB/s]
Downloading pytorch_model.bin: 10%|▉ | 262M/2.63G [00:05<00:43, 54.6MB/s]
Downloading pytorch_model.bin: 10%|█ | 273M/2.63G [00:05<00:43, 54.2MB/s]
Downloading pytorch_model.bin: 11%|█ | 283M/2.63G [00:05<00:43, 53.7MB/s]
Downloading pytorch_model.bin: 11%|█ | 294M/2.63G [00:05<00:43, 53.5MB/s]
Downloading pytorch_model.bin: 12%|█▏ | 304M/2.63G [00:06<00:43, 52.9MB/s]
Downloading pytorch_model.bin: 12%|█▏ | 315M/2.63G [00:06<00:45, 51.3MB/s]
Downloading pytorch_model.bin: 12%|█▏ | 325M/2.63G [00:06<00:44, 51.3MB/s]
Downloading pytorch_model.bin: 13%|█▎ | 336M/2.63G [00:06<00:44, 51.6MB/s]
Downloading pytorch_model.bin: 13%|█▎ | 346M/2.63G [00:06<00:43, 52.3MB/s]
Downloading pytorch_model.bin: 14%|█▎ | 357M/2.63G [00:07<00:43, 52.3MB/s]
Downloading pytorch_model.bin: 14%|█▍ | 367M/2.63G [00:07<00:43, 52.1MB/s]
Downloading pytorch_model.bin: 14%|█▍ | 377M/2.63G [00:07<00:43, 52.2MB/s]
Downloading pytorch_model.bin: 15%|█▍ | 388M/2.63G [00:07<00:43, 51.9MB/s]
Downloading pytorch_model.bin: 15%|█▌ | 398M/2.63G [00:07<00:44, 50.2MB/s]
Downloading pytorch_model.bin: 16%|█▌ | 409M/2.63G [00:08<00:45, 49.0MB/s]
Downloading pytorch_model.bin: 16%|█▌ | 419M/2.63G [00:08<00:44, 49.8MB/s]
Downloading pytorch_model.bin: 16%|█▋ | 430M/2.63G [00:08<00:42, 52.0MB/s]
Downloading pytorch_model.bin: 17%|█▋ | 440M/2.63G [00:08<00:39, 55.4MB/s]
Downloading pytorch_model.bin: 17%|█▋ | 451M/2.63G [00:08<00:38, 57.3MB/s]
Downloading pytorch_model.bin: 18%|█▊ | 461M/2.63G [00:09<00:38, 56.7MB/s]
Downloading pytorch_model.bin: 18%|█▊ | 472M/2.63G [00:09<00:38, 55.7MB/s]
Downloading pytorch_model.bin: 18%|█▊ | 482M/2.63G [00:09<00:38, 55.4MB/s]
Downloading pytorch_model.bin: 19%|█▊ | 493M/2.63G [00:09<00:39, 53.5MB/s]
Downloading pytorch_model.bin: 19%|█▉ | 503M/2.63G [00:09<00:39, 54.3MB/s]
Downloading pytorch_model.bin: 20%|█▉ | 514M/2.63G [00:10<00:39, 54.0MB/s]
Downloading pytorch_model.bin: 20%|█▉ | 524M/2.63G [00:10<00:39, 53.8MB/s]
Downloading pytorch_model.bin: 20%|██ | 535M/2.63G [00:10<00:39, 53.6MB/s]
Downloading pytorch_model.bin: 21%|██ | 545M/2.63G [00:10<00:39, 53.1MB/s]
Downloading pytorch_model.bin: 21%|██ | 556M/2.63G [00:10<00:39, 52.9MB/s]
Downloading pytorch_model.bin: 22%|██▏ | 566M/2.63G [00:11<00:39, 52.7MB/s]
Downloading pytorch_model.bin: 22%|██▏ | 577M/2.63G [00:11<00:39, 52.5MB/s]
Downloading pytorch_model.bin: 22%|██▏ | 587M/2.63G [00:11<00:38, 52.5MB/s]
Downloading pytorch_model.bin: 23%|██▎ | 598M/2.63G [00:11<00:39, 52.0MB/s]
Downloading pytorch_model.bin: 23%|██▎ | 608M/2.63G [00:11<00:38, 51.9MB/s]
Downloading pytorch_model.bin: 24%|██▎ | 619M/2.63G [00:12<00:38, 51.8MB/s]
Downloading pytorch_model.bin: 24%|██▍ | 629M/2.63G [00:12<00:39, 50.3MB/s]
Downloading pytorch_model.bin: 24%|██▍ | 640M/2.63G [00:12<00:39, 50.1MB/s]
Downloading pytorch_model.bin: 25%|██▍ | 650M/2.63G [00:12<00:39, 50.8MB/s]
Downloading pytorch_model.bin: 25%|██▌ | 661M/2.63G [00:12<00:38, 50.8MB/s]
Downloading pytorch_model.bin: 26%|██▌ | 671M/2.63G [00:13<00:38, 51.3MB/s]
Downloading pytorch_model.bin: 26%|██▌ | 682M/2.63G [00:13<00:37, 51.5MB/s]
Downloading pytorch_model.bin: 26%|██▋ | 692M/2.63G [00:13<00:37, 51.5MB/s]
Downloading pytorch_model.bin: 27%|██▋ | 703M/2.63G [00:13<00:37, 51.7MB/s]
Downloading pytorch_model.bin: 27%|██▋ | 713M/2.63G [00:13<00:36, 52.6MB/s]
Downloading pytorch_model.bin: 27%|██▋ | 724M/2.63G [00:14<00:35, 53.8MB/s]
Downloading pytorch_model.bin: 28%|██▊ | 734M/2.63G [00:14<00:34, 54.9MB/s]
Downloading pytorch_model.bin: 28%|██▊ | 744M/2.63G [00:14<00:33, 55.6MB/s]
Downloading pytorch_model.bin: 29%|██▊ | 755M/2.63G [00:14<00:33, 55.3MB/s]
Downloading pytorch_model.bin: 29%|██▉ | 765M/2.63G [00:14<00:34, 54.0MB/s]
Downloading pytorch_model.bin: 29%|██▉ | 776M/2.63G [00:15<00:34, 54.0MB/s]
Downloading pytorch_model.bin: 30%|██▉ | 786M/2.63G [00:15<00:33, 55.0MB/s]
Downloading pytorch_model.bin: 30%|███ | 797M/2.63G [00:15<00:33, 55.3MB/s]
Downloading pytorch_model.bin: 31%|███ | 807M/2.63G [00:15<00:32, 56.2MB/s]
Downloading pytorch_model.bin: 31%|███ | 818M/2.63G [00:15<00:31, 58.2MB/s]
Downloading pytorch_model.bin: 31%|███▏ | 828M/2.63G [00:15<00:31, 56.8MB/s]
Downloading pytorch_model.bin: 32%|███▏ | 839M/2.63G [00:16<00:32, 56.0MB/s]
Downloading pytorch_model.bin: 32%|███▏ | 849M/2.63G [00:16<00:31, 56.8MB/s]
Downloading pytorch_model.bin: 33%|███▎ | 860M/2.63G [00:16<00:30, 58.2MB/s]
Downloading pytorch_model.bin: 33%|███▎ | 870M/2.63G [00:16<00:30, 58.6MB/s]
Downloading pytorch_model.bin: 33%|███▎ | 881M/2.63G [00:16<00:30, 56.6MB/s]
Downloading pytorch_model.bin: 34%|███▍ | 891M/2.63G [00:17<00:31, 55.5MB/s]
Downloading pytorch_model.bin: 34%|███▍ | 902M/2.63G [00:17<00:31, 55.8MB/s]
Downloading pytorch_model.bin: 35%|███▍ | 912M/2.63G [00:17<00:30, 56.4MB/s]
Downloading pytorch_model.bin: 35%|███▌ | 923M/2.63G [00:17<00:30, 56.0MB/s]
Downloading pytorch_model.bin: 35%|███▌ | 933M/2.63G [00:17<00:30, 55.4MB/s]
Downloading pytorch_model.bin: 36%|███▌ | 944M/2.63G [00:18<00:30, 55.1MB/s]
Downloading pytorch_model.bin: 36%|███▋ | 954M/2.63G [00:18<00:29, 56.3MB/s]
Downloading pytorch_model.bin: 37%|███▋ | 965M/2.63G [00:18<00:28, 57.7MB/s]
Downloading pytorch_model.bin: 37%|███▋ | 975M/2.63G [00:18<00:29, 57.0MB/s]
Downloading pytorch_model.bin: 37%|███▋ | 986M/2.63G [00:18<00:28, 57.0MB/s]
Downloading pytorch_model.bin: 38%|███▊ | 996M/2.63G [00:18<00:27, 59.5MB/s]
Downloading pytorch_model.bin: 38%|███▊ | 1.01G/2.63G [00:19<00:26, 61.6MB/s]
Downloading pytorch_model.bin: 39%|███▊ | 1.02G/2.63G [00:19<00:25, 62.4MB/s]
Downloading pytorch_model.bin: 39%|███▉ | 1.03G/2.63G [00:19<00:25, 63.3MB/s]
Downloading pytorch_model.bin: 39%|███▉ | 1.04G/2.63G [00:19<00:24, 64.3MB/s]
Downloading pytorch_model.bin: 40%|███▉ | 1.05G/2.63G [00:19<00:24, 65.0MB/s]
Downloading pytorch_model.bin: 40%|████ | 1.06G/2.63G [00:19<00:24, 64.6MB/s]
Downloading pytorch_model.bin: 41%|████ | 1.07G/2.63G [00:19<00:23, 65.2MB/s]
Downloading pytorch_model.bin: 41%|████ | 1.08G/2.63G [00:20<00:24, 63.0MB/s]
Downloading pytorch_model.bin: 41%|████▏ | 1.09G/2.63G [00:20<00:25, 59.6MB/s]
Downloading pytorch_model.bin: 42%|████▏ | 1.10G/2.63G [00:20<00:26, 58.8MB/s]
Downloading pytorch_model.bin: 42%|████▏ | 1.11G/2.63G [00:20<00:25, 58.5MB/s]
Downloading pytorch_model.bin: 43%|████▎ | 1.12G/2.63G [00:20<00:26, 57.9MB/s]
Downloading pytorch_model.bin: 43%|████▎ | 1.13G/2.63G [00:21<00:26, 57.5MB/s]
Downloading pytorch_model.bin: 43%|████▎ | 1.14G/2.63G [00:21<00:25, 57.5MB/s]
Downloading pytorch_model.bin: 44%|████▍ | 1.15G/2.63G [00:21<00:26, 55.7MB/s]
Downloading pytorch_model.bin: 44%|████▍ | 1.16G/2.63G [00:21<00:26, 55.5MB/s]
Downloading pytorch_model.bin: 45%|████▍ | 1.17G/2.63G [00:21<00:25, 56.1MB/s]
Downloading pytorch_model.bin: 45%|████▌ | 1.18G/2.63G [00:22<00:25, 56.7MB/s]
Downloading pytorch_model.bin: 45%|████▌ | 1.20G/2.63G [00:22<00:25, 57.1MB/s]
Downloading pytorch_model.bin: 46%|████▌ | 1.21G/2.63G [00:22<00:25, 56.4MB/s]
Downloading pytorch_model.bin: 46%|████▌ | 1.22G/2.63G [00:22<00:24, 57.1MB/s]
Downloading pytorch_model.bin: 47%|████▋ | 1.23G/2.63G [00:22<00:24, 56.6MB/s]
Downloading pytorch_model.bin: 47%|████▋ | 1.24G/2.63G [00:22<00:24, 56.3MB/s]
Downloading pytorch_model.bin: 47%|████▋ | 1.25G/2.63G [00:23<00:23, 58.4MB/s]
Downloading pytorch_model.bin: 48%|████▊ | 1.26G/2.63G [00:23<00:22, 60.6MB/s]
Downloading pytorch_model.bin: 48%|████▊ | 1.27G/2.63G [00:23<00:22, 61.9MB/s]
Downloading pytorch_model.bin: 49%|████▊ | 1.28G/2.63G [00:23<00:21, 63.2MB/s]
Downloading pytorch_model.bin: 49%|████▉ | 1.29G/2.63G [00:23<00:20, 64.3MB/s]
Downloading pytorch_model.bin: 49%|████▉ | 1.30G/2.63G [00:23<00:20, 64.4MB/s]
Downloading pytorch_model.bin: 50%|████▉ | 1.31G/2.63G [00:24<00:20, 64.9MB/s]
Downloading pytorch_model.bin: 50%|█████ | 1.32G/2.63G [00:24<00:20, 64.6MB/s]
Downloading pytorch_model.bin: 51%|█████ | 1.33G/2.63G [00:24<00:21, 61.5MB/s]
Downloading pytorch_model.bin: 51%|█████ | 1.34G/2.63G [00:24<00:21, 59.9MB/s]
Downloading pytorch_model.bin: 51%|█████▏ | 1.35G/2.63G [00:24<00:21, 58.7MB/s]
Downloading pytorch_model.bin: 52%|█████▏ | 1.36G/2.63G [00:25<00:22, 57.6MB/s]
Downloading pytorch_model.bin: 52%|█████▏ | 1.37G/2.63G [00:25<00:22, 56.0MB/s]
Downloading pytorch_model.bin: 53%|█████▎ | 1.38G/2.63G [00:25<00:21, 57.4MB/s]
Downloading pytorch_model.bin: 53%|█████▎ | 1.39G/2.63G [00:25<00:22, 56.2MB/s]
Downloading pytorch_model.bin: 53%|█████▎ | 1.41G/2.63G [00:25<00:22, 55.2MB/s]
Downloading pytorch_model.bin: 54%|█████▍ | 1.42G/2.63G [00:25<00:21, 56.4MB/s]
Downloading pytorch_model.bin: 54%|█████▍ | 1.43G/2.63G [00:26<00:21, 56.6MB/s]
Downloading pytorch_model.bin: 55%|█████▍ | 1.44G/2.63G [00:26<00:21, 56.5MB/s]
Downloading pytorch_model.bin: 55%|█████▍ | 1.45G/2.63G [00:26<00:21, 55.7MB/s]
Downloading pytorch_model.bin: 55%|█████▌ | 1.46G/2.63G [00:26<00:29, 39.8MB/s]
Downloading pytorch_model.bin: 56%|█████▌ | 1.47G/2.63G [00:27<00:26, 43.7MB/s]
Downloading pytorch_model.bin: 56%|█████▌ | 1.48G/2.63G [00:27<00:24, 46.7MB/s]
Downloading pytorch_model.bin: 57%|█████▋ | 1.49G/2.63G [00:27<00:23, 49.3MB/s]
Downloading pytorch_model.bin: 57%|█████▋ | 1.50G/2.63G [00:27<00:22, 51.0MB/s]
Downloading pytorch_model.bin: 57%|█████▋ | 1.51G/2.63G [00:27<00:21, 51.7MB/s]
Downloading pytorch_model.bin: 58%|█████▊ | 1.52G/2.63G [00:28<00:21, 51.9MB/s]
Downloading pytorch_model.bin: 58%|█████▊ | 1.53G/2.63G [00:28<00:20, 54.8MB/s]
Downloading pytorch_model.bin: 59%|█████▊ | 1.54G/2.63G [00:28<00:18, 58.4MB/s]
Downloading pytorch_model.bin: 59%|█████▉ | 1.55G/2.63G [00:28<00:18, 57.4MB/s]
Downloading pytorch_model.bin: 59%|█████▉ | 1.56G/2.63G [00:28<00:18, 58.9MB/s]
Downloading pytorch_model.bin: 60%|█████▉ | 1.57G/2.63G [00:28<00:18, 57.2MB/s]
Downloading pytorch_model.bin: 60%|██████ | 1.58G/2.63G [00:29<00:18, 56.0MB/s]
Downloading pytorch_model.bin: 61%|██████ | 1.59G/2.63G [00:29<00:18, 57.1MB/s]
Downloading pytorch_model.bin: 61%|██████ | 1.60G/2.63G [00:29<00:17, 57.3MB/s]
Downloading pytorch_model.bin: 61%|██████▏ | 1.61G/2.63G [00:29<00:17, 56.5MB/s]
Downloading pytorch_model.bin: 62%|██████▏ | 1.63G/2.63G [00:29<00:18, 55.5MB/s]
Downloading pytorch_model.bin: 62%|██████▏ | 1.64G/2.63G [00:30<00:17, 58.5MB/s]
Downloading pytorch_model.bin: 63%|██████▎ | 1.65G/2.63G [00:30<00:16, 58.1MB/s]
Downloading pytorch_model.bin: 63%|██████▎ | 1.66G/2.63G [00:30<00:17, 57.0MB/s]
Downloading pytorch_model.bin: 63%|██████▎ | 1.67G/2.63G [00:30<00:17, 56.5MB/s]
Downloading pytorch_model.bin: 64%|██████▍ | 1.68G/2.63G [00:30<00:17, 55.6MB/s]
Downloading pytorch_model.bin: 64%|██████▍ | 1.69G/2.63G [00:31<00:17, 55.3MB/s]
Downloading pytorch_model.bin: 65%|██████▍ | 1.70G/2.63G [00:31<00:16, 56.0MB/s]
Downloading pytorch_model.bin: 65%|██████▍ | 1.71G/2.63G [00:31<00:16, 56.6MB/s]
Downloading pytorch_model.bin: 65%|██████▌ | 1.72G/2.63G [00:31<00:16, 56.7MB/s]
Downloading pytorch_model.bin: 66%|██████▌ | 1.73G/2.63G [00:31<00:16, 55.2MB/s]
Downloading pytorch_model.bin: 66%|██████▌ | 1.74G/2.63G [00:31<00:15, 56.0MB/s]
Downloading pytorch_model.bin: 67%|██████▋ | 1.75G/2.63G [00:32<00:15, 56.7MB/s]
Downloading pytorch_model.bin: 67%|██████▋ | 1.76G/2.63G [00:32<00:15, 56.3MB/s]
Downloading pytorch_model.bin: 67%|██████▋ | 1.77G/2.63G [00:32<00:15, 56.8MB/s]
Downloading pytorch_model.bin: 68%|██████▊ | 1.78G/2.63G [00:32<00:15, 56.5MB/s]
Downloading pytorch_model.bin: 68%|██████▊ | 1.79G/2.63G [00:32<00:14, 56.7MB/s]
Downloading pytorch_model.bin: 69%|██████▊ | 1.80G/2.63G [00:33<00:14, 55.3MB/s]
Downloading pytorch_model.bin: 69%|██████▉ | 1.81G/2.63G [00:33<00:14, 55.4MB/s]
Downloading pytorch_model.bin: 69%|██████▉ | 1.82G/2.63G [00:33<00:14, 55.9MB/s]
Downloading pytorch_model.bin: 70%|██████▉ | 1.84G/2.63G [00:33<00:13, 58.7MB/s]
Downloading pytorch_model.bin: 70%|███████ | 1.85G/2.63G [00:33<00:12, 60.7MB/s]
Downloading pytorch_model.bin: 71%|███████ | 1.86G/2.63G [00:33<00:12, 62.2MB/s]
Downloading pytorch_model.bin: 71%|███████ | 1.87G/2.63G [00:34<00:12, 61.3MB/s]
Downloading pytorch_model.bin: 71%|███████▏ | 1.88G/2.63G [00:34<00:12, 60.5MB/s]
Downloading pytorch_model.bin: 72%|███████▏ | 1.89G/2.63G [00:34<00:12, 59.1MB/s]
Downloading pytorch_model.bin: 72%|███████▏ | 1.90G/2.63G [00:34<00:12, 58.6MB/s]
Downloading pytorch_model.bin: 73%|███████▎ | 1.91G/2.63G [00:34<00:12, 59.5MB/s]
Downloading pytorch_model.bin: 73%|███████▎ | 1.92G/2.63G [00:34<00:11, 61.3MB/s]
Downloading pytorch_model.bin: 73%|███████▎ | 1.93G/2.63G [00:35<00:11, 60.6MB/s]
Downloading pytorch_model.bin: 74%|███████▎ | 1.94G/2.63G [00:35<00:10, 63.3MB/s]
Downloading pytorch_model.bin: 74%|███████▍ | 1.95G/2.63G [00:35<00:10, 65.3MB/s]
Downloading pytorch_model.bin: 75%|███████▍ | 1.96G/2.63G [00:35<00:10, 64.7MB/s]
Downloading pytorch_model.bin: 75%|███████▍ | 1.97G/2.63G [00:35<00:10, 62.6MB/s]
Downloading pytorch_model.bin: 75%|███████▌ | 1.98G/2.63G [00:35<00:10, 61.0MB/s]
Downloading pytorch_model.bin: 76%|███████▌ | 1.99G/2.63G [00:36<00:10, 59.1MB/s]
Downloading pytorch_model.bin: 76%|███████▌ | 2.00G/2.63G [00:36<00:11, 56.9MB/s]
Downloading pytorch_model.bin: 77%|███████▋ | 2.01G/2.63G [00:36<00:10, 56.6MB/s]
Downloading pytorch_model.bin: 77%|███████▋ | 2.02G/2.63G [00:36<00:10, 57.2MB/s]
Downloading pytorch_model.bin: 77%|███████▋ | 2.03G/2.63G [00:36<00:10, 57.4MB/s]
Downloading pytorch_model.bin: 78%|███████▊ | 2.04G/2.63G [00:37<00:09, 59.5MB/s]
Downloading pytorch_model.bin: 78%|███████▊ | 2.06G/2.63G [00:37<00:09, 61.3MB/s]
Downloading pytorch_model.bin: 78%|███████▊ | 2.07G/2.63G [00:37<00:09, 62.7MB/s]
Downloading pytorch_model.bin: 79%|███████▉ | 2.08G/2.63G [00:37<00:08, 64.0MB/s]
Downloading pytorch_model.bin: 79%|███████▉ | 2.09G/2.63G [00:37<00:08, 63.6MB/s]
Downloading pytorch_model.bin: 80%|███████▉ | 2.10G/2.63G [00:37<00:08, 64.8MB/s]
Downloading pytorch_model.bin: 80%|████████ | 2.11G/2.63G [00:38<00:08, 65.3MB/s]
Downloading pytorch_model.bin: 80%|████████ | 2.12G/2.63G [00:38<00:07, 64.7MB/s]
Downloading pytorch_model.bin: 81%|████████ | 2.13G/2.63G [00:38<00:07, 65.5MB/s]
Downloading pytorch_model.bin: 81%|████████▏ | 2.14G/2.63G [00:38<00:11, 42.7MB/s]
Downloading pytorch_model.bin: 82%|████████▏ | 2.15G/2.63G [00:38<00:10, 47.3MB/s]
Downloading pytorch_model.bin: 82%|████████▏ | 2.16G/2.63G [00:39<00:09, 50.6MB/s]
Downloading pytorch_model.bin: 82%|████████▏ | 2.17G/2.63G [00:39<00:08, 52.0MB/s]
Downloading pytorch_model.bin: 83%|████████▎ | 2.18G/2.63G [00:39<00:08, 52.6MB/s]
Downloading pytorch_model.bin: 83%|████████▎ | 2.19G/2.63G [00:39<00:08, 52.7MB/s]
Downloading pytorch_model.bin: 84%|████████▎ | 2.20G/2.63G [00:39<00:08, 53.2MB/s]
Downloading pytorch_model.bin: 84%|████████▍ | 2.21G/2.63G [00:40<00:07, 53.2MB/s]
Downloading pytorch_model.bin: 84%|████████▍ | 2.22G/2.63G [00:40<00:07, 53.3MB/s]
Downloading pytorch_model.bin: 85%|████████▍ | 2.23G/2.63G [00:40<00:07, 54.2MB/s]
Downloading pytorch_model.bin: 85%|████████▌ | 2.24G/2.63G [00:40<00:07, 54.7MB/s]
Downloading pytorch_model.bin: 86%|████████▌ | 2.25G/2.63G [00:40<00:06, 55.2MB/s]
Downloading pytorch_model.bin: 86%|████████▌ | 2.26G/2.63G [00:41<00:06, 54.2MB/s]
Downloading pytorch_model.bin: 86%|████████▋ | 2.28G/2.63G [00:41<00:06, 54.6MB/s]
Downloading pytorch_model.bin: 87%|████████▋ | 2.29G/2.63G [00:41<00:06, 54.8MB/s]
Downloading pytorch_model.bin: 87%|████████▋ | 2.30G/2.63G [00:41<00:06, 54.6MB/s]
Downloading pytorch_model.bin: 88%|████████▊ | 2.31G/2.63G [00:41<00:05, 54.7MB/s]
Downloading pytorch_model.bin: 88%|████████▊ | 2.32G/2.63G [00:42<00:05, 54.8MB/s]
Downloading pytorch_model.bin: 88%|████████▊ | 2.33G/2.63G [00:42<00:05, 55.0MB/s]
Downloading pytorch_model.bin: 89%|████████▉ | 2.34G/2.63G [00:42<00:05, 55.3MB/s]
Downloading pytorch_model.bin: 89%|████████▉ | 2.35G/2.63G [00:42<00:05, 55.4MB/s]
Downloading pytorch_model.bin: 90%|████████▉ | 2.36G/2.63G [00:42<00:04, 55.8MB/s]
Downloading pytorch_model.bin: 90%|█████████ | 2.37G/2.63G [00:42<00:04, 55.6MB/s]
Downloading pytorch_model.bin: 90%|█████████ | 2.38G/2.63G [00:43<00:04, 55.3MB/s]
Downloading pytorch_model.bin: 91%|█████████ | 2.39G/2.63G [00:43<00:04, 54.9MB/s]
Downloading pytorch_model.bin: 91%|█████████ | 2.40G/2.63G [00:43<00:04, 55.8MB/s]
Downloading pytorch_model.bin: 92%|█████████▏| 2.41G/2.63G [00:43<00:03, 58.8MB/s]
Downloading pytorch_model.bin: 92%|█████████▏| 2.42G/2.63G [00:43<00:03, 60.0MB/s]
Downloading pytorch_model.bin: 92%|█████████▏| 2.43G/2.63G [00:44<00:03, 61.9MB/s]
Downloading pytorch_model.bin: 93%|█████████▎| 2.44G/2.63G [00:44<00:03, 58.7MB/s]
Downloading pytorch_model.bin: 93%|█████████▎| 2.45G/2.63G [00:44<00:03, 57.4MB/s]
Downloading pytorch_model.bin: 94%|█████████▎| 2.46G/2.63G [00:44<00:02, 59.4MB/s]
Downloading pytorch_model.bin: 94%|█████████▍| 2.47G/2.63G [00:44<00:02, 58.3MB/s]
Downloading pytorch_model.bin: 94%|█████████▍| 2.49G/2.63G [00:44<00:02, 58.3MB/s]
Downloading pytorch_model.bin: 95%|█████████▍| 2.50G/2.63G [00:45<00:02, 58.2MB/s]
Downloading pytorch_model.bin: 95%|█████████▌| 2.51G/2.63G [00:45<00:02, 57.6MB/s]
Downloading pytorch_model.bin: 96%|█████████▌| 2.52G/2.63G [00:45<00:01, 57.9MB/s]
Downloading pytorch_model.bin: 96%|█████████▌| 2.53G/2.63G [00:45<00:01, 57.1MB/s]
Downloading pytorch_model.bin: 96%|█████████▋| 2.54G/2.63G [00:45<00:01, 57.1MB/s]
Downloading pytorch_model.bin: 97%|█████████▋| 2.55G/2.63G [00:46<00:01, 55.8MB/s]
Downloading pytorch_model.bin: 97%|█████████▋| 2.56G/2.63G [00:46<00:01, 56.4MB/s]
Downloading pytorch_model.bin: 98%|█████████▊| 2.57G/2.63G [00:46<00:01, 56.7MB/s]
Downloading pytorch_model.bin: 98%|█████████▊| 2.58G/2.63G [00:46<00:00, 56.9MB/s]
Downloading pytorch_model.bin: 98%|█████████▊| 2.59G/2.63G [00:46<00:00, 56.4MB/s]
Downloading pytorch_model.bin: 99%|█████████▉| 2.60G/2.63G [00:46<00:00, 56.2MB/s]
Downloading pytorch_model.bin: 99%|█████████▉| 2.61G/2.63G [00:47<00:00, 56.5MB/s]
Downloading pytorch_model.bin: 100%|█████████▉| 2.62G/2.63G [00:47<00:00, 56.6MB/s]
Downloading pytorch_model.bin: 100%|██████████| 2.63G/2.63G [00:47<00:00, 55.6MB/s]
Downloading pytorch_model.bin: 100%|██████████| 2.63G/2.63G [00:47<00:00, 55.4MB/s]
Downloading (…)neration_config.json: 0%| | 0.00/137 [00:00<?, ?B/s]
Downloading (…)neration_config.json: 100%|██████████| 137/137 [00:00<00:00, 8.19kB/s]
Downloading metadata: 0%| | 0.00/926 [00:00<?, ?B/s]
Downloading metadata: 100%|██████████| 926/926 [00:00<00:00, 391kB/s]
Downloading readme: 0%| | 0.00/530 [00:00<?, ?B/s]
Downloading readme: 100%|██████████| 530/530 [00:00<00:00, 242kB/s]
Downloading and preparing dataset None/None to /root/.cache/huggingface/datasets/Dahoas___parquet/default-b9d2c4937d617106/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec...
Downloading data files: 0%| | 0/2 [00:00<?, ?it/s]
Downloading data: 0%| | 0.00/4.61M [00:00<?, ?B/s][A
Downloading data: 2%|▏ | 97.3k/4.61M [00:00<00:04, 912kB/s][A
Downloading data: 12%|█▏ | 574k/4.61M [00:00<00:01, 3.12MB/s][A
Downloading data: 25%|██▌ | 1.16M/4.61M [00:00<00:00, 3.61MB/s][A
Downloading data: 33%|███▎ | 1.54M/4.61M [00:00<00:01, 2.41MB/s][A
Downloading data: 79%|███████▉ | 3.64M/4.61M [00:00<00:00, 6.94MB/s][A
Downloading data: 99%|█████████▊| 4.55M/4.61M [00:00<00:00, 6.60MB/s][A
Downloading data: 100%|██████████| 4.61M/4.61M [00:00<00:00, 5.31MB/s]
Downloading data files: 50%|█████ | 1/2 [00:02<00:02, 2.85s/it]
Downloading data: 0%| | 0.00/68.4M [00:00<?, ?B/s][A
Downloading data: 0%| | 82.9k/68.4M [00:00<01:22, 825kB/s][A
Downloading data: 1%| | 540k/68.4M [00:00<00:22, 3.01MB/s][A
Downloading data: 4%|▍ | 2.84M/68.4M [00:00<00:05, 12.1MB/s][A
Downloading data: 14%|█▍ | 9.52M/68.4M [00:00<00:01, 33.7MB/s][A
Downloading data: 24%|██▍ | 16.7M/68.4M [00:00<00:01, 47.4MB/s][A
Downloading data: 34%|███▍ | 23.5M/68.4M [00:00<00:00, 54.3MB/s][A
Downloading data: 45%|████▌ | 30.8M/68.4M [00:00<00:00, 60.5MB/s][A
Downloading data: 55%|█████▌ | 37.9M/68.4M [00:00<00:00, 63.8MB/s][A
Downloading data: 66%|██████▌ | 45.1M/68.4M [00:00<00:00, 66.4MB/s][A
Downloading data: 76%|███████▋ | 52.3M/68.4M [00:01<00:00, 68.0MB/s][A
Downloading data: 87%|████████▋ | 59.5M/68.4M [00:01<00:00, 69.3MB/s][A
Downloading data: 97%|█████████▋| 66.6M/68.4M [00:01<00:00, 69.8MB/s][A
Downloading data: 100%|██████████| 68.4M/68.4M [00:01<00:00, 55.6MB/s]
Downloading data files: 100%|██████████| 2/2 [00:06<00:00, 3.07s/it]
Downloading data files: 100%|██████████| 2/2 [00:06<00:00, 3.04s/it]
Extracting data files: 0%| | 0/2 [00:00<?, ?it/s]
Extracting data files: 100%|██████████| 2/2 [00:00<00:00, 1741.46it/s]
Generating test split: 0%| | 0/5103 [00:00<?, ? examples/s]
Generating train split: 0%| | 0/76256 [00:00<?, ? examples/s]
Generating train split: 52%|█████▏ | 40000/76256 [00:00<00:00, 249465.69 examples/s]
Generating train split: 92%|█████████▏| 70000/76256 [00:00<00:00, 223221.40 examples/s]
Dataset parquet downloaded and prepared to /root/.cache/huggingface/datasets/Dahoas___parquet/default-b9d2c4937d617106/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec. Subsequent calls will reuse this data.
0%| | 0/2 [00:00<?, ?it/s]
100%|██████████| 2/2 [00:00<00:00, 208.87it/s]
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Using /root/.cache/torch_extensions/py38_cu117 as PyTorch extensions root...
Creating extension directory /root/.cache/torch_extensions/py38_cu117/fused_adam...
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Detected CUDA files, patching ldflags
Emitting ninja build file /root/.cache/torch_extensions/py38_cu117/fused_adam/build.ninja...
Building extension module fused_adam...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
[1/3] /usr/local/cuda/bin/nvcc -DTORCH_EXTENSION_NAME=fused_adam -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -I/usr/local/lib/python3.8/site-packages/deepspeed/ops/csrc/includes -I/usr/local/lib/python3.8/site-packages/deepspeed/ops/csrc/adam -isystem /usr/local/lib/python3.8/site-packages/torch/include -isystem /usr/local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /usr/local/lib/python3.8/site-packages/torch/include/TH -isystem /usr/local/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda/include -isystem /usr/local/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_80,code=compute_80 -gencode=arch=compute_80,code=sm_80 --compiler-options '-fPIC' -O3 -DVERSION_GE_1_1 -DVERSION_GE_1_3 -DVERSION_GE_1_5 -lineinfo --use_fast_math -gencode=arch=compute_80,code=sm_80 -gencode=arch=compute_80,code=compute_80 -std=c++14 -c /usr/local/lib/python3.8/site-packages/deepspeed/ops/csrc/adam/multi_tensor_adam.cu -o multi_tensor_adam.cuda.o
[2/3] c++ -MMD -MF fused_adam_frontend.o.d -DTORCH_EXTENSION_NAME=fused_adam -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -I/usr/local/lib/python3.8/site-packages/deepspeed/ops/csrc/includes -I/usr/local/lib/python3.8/site-packages/deepspeed/ops/csrc/adam -isystem /usr/local/lib/python3.8/site-packages/torch/include -isystem /usr/local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /usr/local/lib/python3.8/site-packages/torch/include/TH -isystem /usr/local/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda/include -isystem /usr/local/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++14 -O3 -std=c++14 -g -Wno-reorder -DVERSION_GE_1_1 -DVERSION_GE_1_3 -DVERSION_GE_1_5 -c /usr/local/lib/python3.8/site-packages/deepspeed/ops/csrc/adam/fused_adam_frontend.cpp -o fused_adam_frontend.o
[3/3] c++ fused_adam_frontend.o multi_tensor_adam.cuda.o -shared -L/usr/local/lib/python3.8/site-packages/torch/lib -lc10 -lc10_cuda -ltorch_cpu -ltorch_cuda_cu -ltorch_cuda_cpp -ltorch -ltorch_python -L/usr/local/cuda/lib64 -lcudart -o fused_adam.so
Loading extension module fused_adam...
Time to load fused_adam op: 21.613110065460205 seconds
[2023-04-14 07:00:27,708] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed info: version=0.9.0, git-hash=unknown, git-branch=unknown
[2023-04-14 07:00:27,713] [INFO] [comm.py:580:init_distributed] Distributed backend already initialized
[2023-04-14 07:00:29,668] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Flops Profiler Enabled: False
[2023-04-14 07:00:29,669] [INFO] [logging.py:96:log_dist] [Rank 0] Removing param_group that has no 'params' in the client Optimizer
[2023-04-14 07:00:29,669] [INFO] [logging.py:96:log_dist] [Rank 0] Using client Optimizer as basic optimizer
[2023-04-14 07:00:29,695] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Basic Optimizer = FusedAdam
[2023-04-14 07:00:29,695] [INFO] [logging.py:96:log_dist] [Rank 0] Creating fp16 optimizer with dynamic loss scale
[2023-04-14 07:00:29,717] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Final Optimizer = FusedAdam
[2023-04-14 07:00:29,717] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed using client LR scheduler
[2023-04-14 07:00:29,717] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed LR Scheduler = <torch.optim.lr_scheduler.LambdaLR object at 0x7fd0a0fe5310>
[2023-04-14 07:00:29,718] [INFO] [logging.py:96:log_dist] [Rank 0] step=0, skipped=0, lr=[0.001, 0.001], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:00:29,718] [INFO] [config.py:953:print] DeepSpeedEngine configuration:
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] activation_checkpointing_config {
"partition_activations": false,
"contiguous_memory_optimization": false,
"cpu_checkpointing": false,
"number_checkpoints": null,
"synchronize_checkpoint_boundary": false,
"profile": false
}
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True}
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] amp_enabled .................. False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] amp_params ................... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] autotuning_config ............ {
"enabled": false,
"start_step": null,
"end_step": null,
"metric_path": null,
"arg_mappings": null,
"metric": "throughput",
"model_info": null,
"results_dir": "autotuning_results",
"exps_dir": "autotuning_exps",
"overwrite": true,
"fast": true,
"start_profile_step": 3,
"end_profile_step": 5,
"tuner_type": "gridsearch",
"tuner_early_stopping": 5,
"tuner_num_trials": 50,
"model_info_path": null,
"mp_size": 1,
"max_train_batch_size": null,
"min_train_batch_size": 1,
"max_train_micro_batch_size_per_gpu": 1.024000e+03,
"min_train_micro_batch_size_per_gpu": 1,
"num_tuning_micro_batch_sizes": 3
}
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] bfloat16_enabled ............. False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] checkpoint_parallel_write_pipeline False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] checkpoint_tag_validation_enabled True
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] checkpoint_tag_validation_fail False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] comms_config ................. <deepspeed.comm.config.DeepSpeedCommsConfig object at 0x7fd010dcb790>
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] communication_data_type ...... None
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] compression_config ........... {'weight_quantization': {'shared_parameters': {'enabled': False, 'quantizer_kernel': False, 'schedule_offset': 0, 'quantize_groups': 1, 'quantize_verbose': False, 'quantization_type': 'symmetric', 'quantize_weight_in_forward': False, 'rounding': 'nearest', 'fp16_mixed_quantize': False, 'quantize_change_ratio': 0.001}, 'different_groups': {}}, 'activation_quantization': {'shared_parameters': {'enabled': False, 'quantization_type': 'symmetric', 'range_calibration': 'dynamic', 'schedule_offset': 1000}, 'different_groups': {}}, 'sparse_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'row_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'head_pruning': {'shared_parameters': {'enabled': False, 'method': 'topk', 'schedule_offset': 1000}, 'different_groups': {}}, 'channel_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'layer_reduction': {'enabled': False}}
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] curriculum_enabled_legacy .... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] curriculum_params_legacy ..... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] data_efficiency_config ....... {'enabled': False, 'seed': 1234, 'data_sampling': {'enabled': False, 'num_epochs': 1000, 'num_workers': 0, 'curriculum_learning': {'enabled': False}}, 'data_routing': {'enabled': False, 'random_ltd': {'enabled': False, 'layer_token_lr_schedule': {'enabled': False}}}}
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] data_efficiency_enabled ...... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] dataloader_drop_last ......... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] disable_allgather ............ False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] dump_state ................... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] dynamic_loss_scale_args ...... {'init_scale': 65536, 'scale_window': 100, 'delayed_shift': 2, 'min_scale': 1}
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_enabled ........... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_gas_boundary_resolution 1
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_layer_name ........ bert.encoder.layer
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_layer_num ......... 0
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_max_iter .......... 100
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_stability ......... 1e-06
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_tol ............... 0.01
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] eigenvalue_verbose ........... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] elasticity_enabled ........... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] flops_profiler_config ........ {
"enabled": false,
"profile_step": 1,
"module_depth": -1,
"top_modules": 1,
"detailed": true,
"output_file": null
}
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] fp16_auto_cast ............... False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] fp16_enabled ................. True
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] fp16_master_weights_and_gradients False
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] global_rank .................. 0
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] grad_accum_dtype ............. None
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] gradient_accumulation_steps .. 2
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] gradient_clipping ............ 1.0
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] gradient_predivide_factor .... 1.0
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] hybrid_engine ................ enabled=False max_out_tokens=512 inference_tp_size=1 release_inference_cache=False pin_parameters=True tp_gather_partition_size=8
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] initial_dynamic_scale ........ 65536
[2023-04-14 07:00:29,719] [INFO] [config.py:957:print] load_universal_checkpoint .... False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] loss_scale ................... 0
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] memory_breakdown ............. False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] monitor_config ............... tensorboard=TensorBoardConfig(enabled=False, output_path='', job_name='DeepSpeedJobName') wandb=WandbConfig(enabled=False, group=None, team=None, project='deepspeed') csv_monitor=CSVConfig(enabled=False, output_path='', job_name='DeepSpeedJobName') enabled=False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] nebula_config ................ {
"enabled": false,
"persistent_storage_path": null,
"persistent_time_interval": 100,
"num_of_version_in_retention": 2,
"enable_nebula_load": true,
"load_path": null
}
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] optimizer_legacy_fusion ...... False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] optimizer_name ............... None
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] optimizer_params ............. None
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0}
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] pld_enabled .................. False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] pld_params ................... False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] prescale_gradients ........... False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] scheduler_name ............... None
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] scheduler_params ............. None
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] sparse_attention ............. None
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] sparse_gradients_enabled ..... False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] steps_per_print .............. 10
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] train_batch_size ............. 32
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] train_micro_batch_size_per_gpu 16
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] use_node_local_storage ....... False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] wall_clock_breakdown ......... False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] world_size ................... 1
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] zero_allow_untested_optimizer False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] zero_config .................. stage=0 contiguous_gradients=True reduce_scatter=True reduce_bucket_size=500,000,000 allgather_partitions=True allgather_bucket_size=500,000,000 overlap_comm=False load_from_fp32_weights=True elastic_checkpoint=False offload_param=DeepSpeedZeroOffloadParamConfig(device='none', nvme_path=None, buffer_count=5, buffer_size=100,000,000, max_in_cpu=1,000,000,000, pin_memory=False) offload_optimizer=DeepSpeedZeroOffloadOptimizerConfig(device='none', nvme_path=None, buffer_count=4, pin_memory=False, pipeline=False, pipeline_read=False, pipeline_write=False, fast_init=False) sub_group_size=1,000,000,000 cpu_offload_param=None cpu_offload_use_pin_memory=None cpu_offload=None prefetch_bucket_size=30000000 param_persistence_threshold=10000 model_persistence_threshold=sys.maxsize max_live_parameters=30000000 max_reuse_distance=1,000,000,000 gather_16bit_weights_on_model_save=False stage3_gather_fp16_weights_on_model_save=False ignore_unused_parameters=True legacy_stage1=False round_robin_gradients=False memory_efficient_linear=False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] zero_enabled ................. False
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] zero_force_ds_cpu_optimizer .. True
[2023-04-14 07:00:29,720] [INFO] [config.py:957:print] zero_optimization_stage ...... 0
[2023-04-14 07:00:29,720] [INFO] [config.py:943:print_user_config] json = {
"train_batch_size": 32,
"train_micro_batch_size_per_gpu": 16,
"steps_per_print": 10,
"zero_optimization": {
"stage": 0,
"offload_param": {
"device": "none"
},
"offload_optimizer": {
"device": "none"
},
"stage3_param_persistence_threshold": 1.000000e+04,
"stage3_max_live_parameters": 3.000000e+07,
"stage3_prefetch_bucket_size": 3.000000e+07,
"memory_efficient_linear": false
},
"fp16": {
"enabled": true,
"loss_scale_window": 100
},
"gradient_clipping": 1.0,
"prescale_gradients": false,
"wall_clock_breakdown": false,
"hybrid_engine": {
"enabled": false,
"inference_tp_size": 1,
"release_inference_cache": false,
"pin_parameters": true,
"tp_gather_partition_size": 8
}
}
Using /root/.cache/torch_extensions/py38_cu117 as PyTorch extensions root...
Creating extension directory /root/.cache/torch_extensions/py38_cu117/utils...
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Emitting ninja build file /root/.cache/torch_extensions/py38_cu117/utils/build.ninja...
Building extension module utils...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
[1/2] c++ -MMD -MF flatten_unflatten.o.d -DTORCH_EXTENSION_NAME=utils -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /usr/local/lib/python3.8/site-packages/torch/include -isystem /usr/local/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /usr/local/lib/python3.8/site-packages/torch/include/TH -isystem /usr/local/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++14 -c /usr/local/lib/python3.8/site-packages/deepspeed/ops/csrc/utils/flatten_unflatten.cpp -o flatten_unflatten.o
[2/2] c++ flatten_unflatten.o -shared -L/usr/local/lib/python3.8/site-packages/torch/lib -lc10 -ltorch_cpu -ltorch -ltorch_python -o utils.so
Loading extension module utils...
Time to load utils op: 12.565346956253052 seconds
***** Running training *****
***** Evaluating perplexity, Epoch 0/1 *****
ppl: 4841.33251953125
Beginning of Epoch 1/1, Total Micro Batches 2860
[2023-04-14 07:01:26,912] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 0
[2023-04-14 07:01:26,912] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 65536 to 32768.0
[2023-04-14 07:01:26,912] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 65536, reducing to 32768.0
[2023-04-14 07:01:28,022] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 1
[2023-04-14 07:01:28,023] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 32768.0 to 16384.0
[2023-04-14 07:01:28,023] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 32768.0, reducing to 16384.0
[2023-04-14 07:01:29,132] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 2
[2023-04-14 07:01:29,133] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 16384.0 to 8192.0
[2023-04-14 07:01:29,133] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 16384.0, reducing to 8192.0
[2023-04-14 07:01:30,237] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 3
[2023-04-14 07:01:30,238] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 8192.0 to 4096.0
[2023-04-14 07:01:30,238] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 8192.0, reducing to 4096.0
[2023-04-14 07:01:37,134] [INFO] [logging.py:96:log_dist] [Rank 0] step=10, skipped=4, lr=[0.0009999565625930518, 0.0009999565625930518], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:01:37,138] [INFO] [timer.py:199:stop] epoch=0/micro_step=20/global_step=10, RunningAvgSamplesPerSec=28.141096377600697, CurrSamplesPerSec=28.12801123295279, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:01:48,572] [INFO] [logging.py:96:log_dist] [Rank 0] step=20, skipped=4, lr=[0.000999691139103864, 0.000999691139103864], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:01:48,576] [INFO] [timer.py:199:stop] epoch=0/micro_step=40/global_step=20, RunningAvgSamplesPerSec=28.078686785579933, CurrSamplesPerSec=28.071358760840397, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:02:00,002] [INFO] [logging.py:96:log_dist] [Rank 0] step=30, skipped=4, lr=[0.0009991845519630679, 0.0009991845519630679], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:02:00,006] [INFO] [timer.py:199:stop] epoch=0/micro_step=60/global_step=30, RunningAvgSamplesPerSec=28.067548117432807, CurrSamplesPerSec=28.11579075481346, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:02:11,418] [INFO] [logging.py:96:log_dist] [Rank 0] step=40, skipped=4, lr=[0.0009984370456625003, 0.0009984370456625003], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:02:11,423] [INFO] [timer.py:199:stop] epoch=0/micro_step=80/global_step=40, RunningAvgSamplesPerSec=28.073548066802058, CurrSamplesPerSec=28.17129833011603, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:02:22,838] [INFO] [logging.py:96:log_dist] [Rank 0] step=50, skipped=4, lr=[0.0009974489809677126, 0.0009974489809677126], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:02:22,842] [INFO] [timer.py:199:stop] epoch=0/micro_step=100/global_step=50, RunningAvgSamplesPerSec=28.073460515306376, CurrSamplesPerSec=28.103057175257835, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:02:34,274] [INFO] [logging.py:96:log_dist] [Rank 0] step=60, skipped=4, lr=[0.0009962208347438538, 0.0009962208347438538], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:02:34,278] [INFO] [timer.py:199:stop] epoch=0/micro_step=120/global_step=60, RunningAvgSamplesPerSec=28.06668333658504, CurrSamplesPerSec=27.944933682231035, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:02:45,732] [INFO] [logging.py:96:log_dist] [Rank 0] step=70, skipped=4, lr=[0.0009947531997255255, 0.0009947531997255255], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:02:45,734] [INFO] [timer.py:199:stop] epoch=0/micro_step=140/global_step=70, RunningAvgSamplesPerSec=28.054937933989542, CurrSamplesPerSec=27.874315487266735, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:02:57,224] [INFO] [logging.py:96:log_dist] [Rank 0] step=80, skipped=4, lr=[0.0009930467842307117, 0.0009930467842307117], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:02:57,227] [INFO] [timer.py:199:stop] epoch=0/micro_step=160/global_step=80, RunningAvgSamplesPerSec=28.03502976436525, CurrSamplesPerSec=27.943706074371658, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:03:08,725] [INFO] [logging.py:96:log_dist] [Rank 0] step=90, skipped=4, lr=[0.0009911024118189266, 0.0009911024118189266], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:03:08,728] [INFO] [timer.py:199:stop] epoch=0/micro_step=180/global_step=90, RunningAvgSamplesPerSec=28.01790135317418, CurrSamplesPerSec=27.889363254819838, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:03:20,199] [INFO] [logging.py:96:log_dist] [Rank 0] step=100, skipped=4, lr=[0.0009889210208937447, 0.0009889210208937447], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:03:20,202] [INFO] [timer.py:199:stop] epoch=0/micro_step=200/global_step=100, RunningAvgSamplesPerSec=28.01027508003987, CurrSamplesPerSec=28.008906574044975, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:03:25,943] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:03:25,943] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 4096.0 to 8192.0
[2023-04-14 07:03:31,714] [INFO] [logging.py:96:log_dist] [Rank 0] step=110, skipped=4, lr=[0.000986503664249902, 0.000986503664249902], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:03:31,717] [INFO] [timer.py:199:stop] epoch=0/micro_step=220/global_step=110, RunningAvgSamplesPerSec=27.994640314629194, CurrSamplesPerSec=27.73325619189209, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:03:43,229] [INFO] [logging.py:96:log_dist] [Rank 0] step=120, skipped=4, lr=[0.000983851508565192, 0.000983851508565192], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:03:43,232] [INFO] [timer.py:199:stop] epoch=0/micro_step=240/global_step=120, RunningAvgSamplesPerSec=27.98217971508587, CurrSamplesPerSec=27.775588813338135, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:03:54,753] [INFO] [logging.py:96:log_dist] [Rank 0] step=130, skipped=4, lr=[0.0009809658338373964, 0.0009809658338373964], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:03:54,756] [INFO] [timer.py:199:stop] epoch=0/micro_step=260/global_step=130, RunningAvgSamplesPerSec=27.969826533526934, CurrSamplesPerSec=27.816915092931872, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:04:06,264] [INFO] [logging.py:96:log_dist] [Rank 0] step=140, skipped=4, lr=[0.0009778480327665255, 0.0009778480327665255], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:04:06,267] [INFO] [timer.py:199:stop] epoch=0/micro_step=280/global_step=140, RunningAvgSamplesPerSec=27.961798295095445, CurrSamplesPerSec=27.948134115000336, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:04:17,780] [INFO] [logging.py:96:log_dist] [Rank 0] step=150, skipped=4, lr=[0.0009744996100826668, 0.0009744996100826668], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:04:17,783] [INFO] [timer.py:199:stop] epoch=0/micro_step=300/global_step=150, RunningAvgSamplesPerSec=27.953563434154024, CurrSamplesPerSec=27.900500251008975, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:04:29,304] [INFO] [logging.py:96:log_dist] [Rank 0] step=160, skipped=4, lr=[0.0009709221818197624, 0.0009709221818197624], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:04:29,307] [INFO] [timer.py:199:stop] epoch=0/micro_step=320/global_step=160, RunningAvgSamplesPerSec=27.94473140562776, CurrSamplesPerSec=27.869627240021313, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:04:40,823] [INFO] [logging.py:96:log_dist] [Rank 0] step=170, skipped=4, lr=[0.0009671174745356714, 0.0009671174745356714], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:04:40,826] [INFO] [timer.py:199:stop] epoch=0/micro_step=340/global_step=170, RunningAvgSamplesPerSec=27.938436426576004, CurrSamplesPerSec=27.914235169869585, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:04:52,287] [INFO] [logging.py:96:log_dist] [Rank 0] step=180, skipped=4, lr=[0.0009630873244788883, 0.0009630873244788883], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:04:52,291] [INFO] [timer.py:199:stop] epoch=0/micro_step=360/global_step=180, RunningAvgSamplesPerSec=27.93923052396758, CurrSamplesPerSec=28.012267836208732, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:05:03,761] [INFO] [logging.py:96:log_dist] [Rank 0] step=190, skipped=4, lr=[0.0009588336767023232, 0.0009588336767023232], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:05:03,765] [INFO] [timer.py:199:stop] epoch=0/micro_step=380/global_step=190, RunningAvgSamplesPerSec=27.939053853941257, CurrSamplesPerSec=27.900796044680277, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:05:15,227] [INFO] [logging.py:96:log_dist] [Rank 0] step=200, skipped=4, lr=[0.0009543585841245694, 0.0009543585841245694], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:05:15,231] [INFO] [timer.py:199:stop] epoch=0/micro_step=400/global_step=200, RunningAvgSamplesPerSec=27.940316446938077, CurrSamplesPerSec=27.99571484352838, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:05:20,945] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:05:20,945] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 8192.0 to 16384.0
[2023-04-14 07:05:26,700] [INFO] [logging.py:96:log_dist] [Rank 0] step=210, skipped=4, lr=[0.0009496642065391134, 0.0009496642065391134], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:05:26,703] [INFO] [timer.py:199:stop] epoch=0/micro_step=420/global_step=210, RunningAvgSamplesPerSec=27.9406361563072, CurrSamplesPerSec=27.793493868223226, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:05:38,150] [INFO] [logging.py:96:log_dist] [Rank 0] step=220, skipped=4, lr=[0.0009447528095719625, 0.0009447528095719625], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:05:38,154] [INFO] [timer.py:199:stop] epoch=0/micro_step=440/global_step=220, RunningAvgSamplesPerSec=27.94290479002408, CurrSamplesPerSec=28.06016123471817, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:05:49,608] [INFO] [logging.py:96:log_dist] [Rank 0] step=230, skipped=4, lr=[0.0009396267635881972, 0.0009396267635881972], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:05:49,612] [INFO] [timer.py:199:stop] epoch=0/micro_step=460/global_step=230, RunningAvgSamplesPerSec=27.943999646258902, CurrSamplesPerSec=27.976067155370266, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:06:01,058] [INFO] [logging.py:96:log_dist] [Rank 0] step=240, skipped=4, lr=[0.0009342885425479722, 0.0009342885425479722], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:06:01,062] [INFO] [timer.py:199:stop] epoch=0/micro_step=480/global_step=240, RunningAvgSamplesPerSec=27.945878308209277, CurrSamplesPerSec=27.97900642619156, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:06:12,521] [INFO] [logging.py:96:log_dist] [Rank 0] step=250, skipped=4, lr=[0.0009287407228125202, 0.0009287407228125202], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:06:12,525] [INFO] [timer.py:199:stop] epoch=0/micro_step=500/global_step=250, RunningAvgSamplesPerSec=27.94650874790295, CurrSamplesPerSec=27.93969237507434, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:06:23,989] [INFO] [logging.py:96:log_dist] [Rank 0] step=260, skipped=4, lr=[0.0009229859819007346, 0.0009229859819007346], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:06:23,994] [INFO] [timer.py:199:stop] epoch=0/micro_step=520/global_step=260, RunningAvgSamplesPerSec=27.946762620632057, CurrSamplesPerSec=27.852716360251165, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:06:35,446] [INFO] [logging.py:96:log_dist] [Rank 0] step=270, skipped=4, lr=[0.0009170270971969311, 0.0009170270971969311], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:06:35,450] [INFO] [timer.py:199:stop] epoch=0/micro_step=540/global_step=270, RunningAvgSamplesPerSec=27.947756017554656, CurrSamplesPerSec=27.903829741761506, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:06:46,919] [INFO] [logging.py:96:log_dist] [Rank 0] step=280, skipped=4, lr=[0.0009108669446104109, 0.0009108669446104109], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:06:46,923] [INFO] [timer.py:199:stop] epoch=0/micro_step=560/global_step=280, RunningAvgSamplesPerSec=27.947241583138933, CurrSamplesPerSec=27.948896507994682, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:06:58,387] [INFO] [logging.py:96:log_dist] [Rank 0] step=290, skipped=4, lr=[0.0009045084971874737, 0.0009045084971874737], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:06:58,391] [INFO] [timer.py:199:stop] epoch=0/micro_step=580/global_step=290, RunningAvgSamplesPerSec=27.94705445274327, CurrSamplesPerSec=27.954659449124346, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:07:09,856] [INFO] [logging.py:96:log_dist] [Rank 0] step=300, skipped=4, lr=[0.0008979548236765506, 0.0008979548236765506], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:07:09,860] [INFO] [timer.py:199:stop] epoch=0/micro_step=600/global_step=300, RunningAvgSamplesPerSec=27.94713926621857, CurrSamplesPerSec=27.96594772558736, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:07:15,588] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:07:15,588] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 16384.0 to 32768.0
[2023-04-14 07:07:21,358] [INFO] [logging.py:96:log_dist] [Rank 0] step=310, skipped=4, lr=[0.0008912090870471478, 0.0008912090870471478], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:07:21,362] [INFO] [timer.py:199:stop] epoch=0/micro_step=620/global_step=310, RunningAvgSamplesPerSec=27.944936270242373, CurrSamplesPerSec=27.91026475482345, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:07:32,837] [INFO] [logging.py:96:log_dist] [Rank 0] step=320, skipped=4, lr=[0.0008842745429633161, 0.0008842745429633161], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:07:32,841] [INFO] [timer.py:199:stop] epoch=0/micro_step=640/global_step=320, RunningAvgSamplesPerSec=27.944254036885653, CurrSamplesPerSec=27.86717377393364, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:07:44,321] [INFO] [logging.py:96:log_dist] [Rank 0] step=330, skipped=4, lr=[0.0008771545382123862, 0.0008771545382123862], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:07:44,325] [INFO] [timer.py:199:stop] epoch=0/micro_step=660/global_step=330, RunningAvgSamplesPerSec=27.94347994504524, CurrSamplesPerSec=27.898429870906778, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:07:55,801] [INFO] [logging.py:96:log_dist] [Rank 0] step=340, skipped=4, lr=[0.0008698525090897231, 0.0008698525090897231], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:07:55,806] [INFO] [timer.py:199:stop] epoch=0/micro_step=680/global_step=340, RunningAvgSamplesPerSec=27.94270471004713, CurrSamplesPerSec=27.936912546325544, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:08:07,325] [INFO] [logging.py:96:log_dist] [Rank 0] step=350, skipped=4, lr=[0.0008623719797402826, 0.0008623719797402826], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:08:07,327] [INFO] [timer.py:199:stop] epoch=0/micro_step=700/global_step=350, RunningAvgSamplesPerSec=27.93934291085674, CurrSamplesPerSec=27.81048847374359, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:08:18,851] [INFO] [logging.py:96:log_dist] [Rank 0] step=360, skipped=4, lr=[0.0008547165604577695, 0.0008547165604577695], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:08:18,854] [INFO] [timer.py:199:stop] epoch=0/micro_step=720/global_step=360, RunningAvgSamplesPerSec=27.93575047924138, CurrSamplesPerSec=27.804871217472908, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:08:30,744] [INFO] [logging.py:96:log_dist] [Rank 0] step=370, skipped=4, lr=[0.0008468899459422181, 0.0008468899459422181], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:08:30,749] [INFO] [timer.py:199:stop] epoch=0/micro_step=740/global_step=370, RunningAvgSamplesPerSec=27.90790507995182, CurrSamplesPerSec=27.907189040202667, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:08:42,219] [INFO] [logging.py:96:log_dist] [Rank 0] step=380, skipped=4, lr=[0.0008388959135168359, 0.0008388959135168359], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:08:42,223] [INFO] [timer.py:199:stop] epoch=0/micro_step=760/global_step=380, RunningAvgSamplesPerSec=27.908444913494918, CurrSamplesPerSec=27.887045373418886, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:08:53,681] [INFO] [logging.py:96:log_dist] [Rank 0] step=390, skipped=4, lr=[0.0008307383213049714, 0.0008307383213049714], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:08:53,684] [INFO] [timer.py:199:stop] epoch=0/micro_step=780/global_step=390, RunningAvgSamplesPerSec=27.90980129021459, CurrSamplesPerSec=27.966454688659038, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:09:05,184] [INFO] [logging.py:96:log_dist] [Rank 0] step=400, skipped=4, lr=[0.0008224211063680853, 0.0008224211063680853], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:09:05,187] [INFO] [timer.py:199:stop] epoch=0/micro_step=800/global_step=400, RunningAvgSamplesPerSec=27.90872959774225, CurrSamplesPerSec=27.84752691848682, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:09:10,935] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:09:10,935] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 32768.0 to 65536.0
[2023-04-14 07:09:16,717] [INFO] [logging.py:96:log_dist] [Rank 0] step=410, skipped=4, lr=[0.0008139482828056254, 0.0008139482828056254], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:09:16,720] [INFO] [timer.py:199:stop] epoch=0/micro_step=820/global_step=410, RunningAvgSamplesPerSec=27.90608462810118, CurrSamplesPerSec=27.829552200369534, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:09:28,236] [INFO] [logging.py:96:log_dist] [Rank 0] step=420, skipped=4, lr=[0.0008053239398177191, 0.0008053239398177191], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:09:28,239] [INFO] [timer.py:199:stop] epoch=0/micro_step=840/global_step=420, RunningAvgSamplesPerSec=27.904361435920347, CurrSamplesPerSec=27.818385275198793, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:09:39,767] [INFO] [logging.py:96:log_dist] [Rank 0] step=430, skipped=4, lr=[0.0007965522397316221, 0.0007965522397316221], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:09:39,770] [INFO] [timer.py:199:stop] epoch=0/micro_step=860/global_step=430, RunningAvgSamplesPerSec=27.90219677277752, CurrSamplesPerSec=27.88143190257329, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:09:51,289] [INFO] [logging.py:96:log_dist] [Rank 0] step=440, skipped=4, lr=[0.000787637415992873, 0.000787637415992873], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:09:51,292] [INFO] [timer.py:199:stop] epoch=0/micro_step=880/global_step=440, RunningAvgSamplesPerSec=27.900581766447058, CurrSamplesPerSec=27.8346656745396, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:10:02,815] [INFO] [logging.py:96:log_dist] [Rank 0] step=450, skipped=4, lr=[0.000778583771122125, 0.000778583771122125], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:10:02,818] [INFO] [timer.py:199:stop] epoch=0/micro_step=900/global_step=450, RunningAvgSamplesPerSec=27.89877176742672, CurrSamplesPerSec=27.79556597095809, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:10:14,348] [INFO] [logging.py:96:log_dist] [Rank 0] step=460, skipped=4, lr=[0.0007693956746386408, 0.0007693956746386408], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:10:14,351] [INFO] [timer.py:199:stop] epoch=0/micro_step=920/global_step=460, RunningAvgSamplesPerSec=27.896732463439644, CurrSamplesPerSec=27.784754146720147, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:10:25,872] [INFO] [logging.py:96:log_dist] [Rank 0] step=470, skipped=4, lr=[0.0007600775609514493, 0.0007600775609514493], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:10:25,875] [INFO] [timer.py:199:stop] epoch=0/micro_step=940/global_step=470, RunningAvgSamplesPerSec=27.89516590035489, CurrSamplesPerSec=27.84345993311814, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:10:37,388] [INFO] [logging.py:96:log_dist] [Rank 0] step=480, skipped=4, lr=[0.0007506339272191898, 0.0007506339272191898], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:10:37,391] [INFO] [timer.py:199:stop] epoch=0/micro_step=960/global_step=480, RunningAvgSamplesPerSec=27.894055137267724, CurrSamplesPerSec=27.834573315230116, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:10:48,874] [INFO] [logging.py:96:log_dist] [Rank 0] step=490, skipped=4, lr=[0.0007410693311796666, 0.0007410693311796666], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:10:48,878] [INFO] [timer.py:199:stop] epoch=0/micro_step=980/global_step=490, RunningAvgSamplesPerSec=27.894183330433044, CurrSamplesPerSec=27.907345711163856, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:11:00,353] [INFO] [logging.py:96:log_dist] [Rank 0] step=500, skipped=4, lr=[0.0007313883889501701, 0.0007313883889501701], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:11:00,357] [INFO] [timer.py:199:stop] epoch=0/micro_step=1000/global_step=500, RunningAvgSamplesPerSec=27.894589838673543, CurrSamplesPerSec=27.908755828978105, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:11:06,083] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:11:06,083] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:11:07,202] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 505
[2023-04-14 07:11:07,202] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:11:07,202] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:11:11,809] [INFO] [logging.py:96:log_dist] [Rank 0] step=510, skipped=5, lr=[0.0007225799254574904, 0.0007225799254574904], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:11:11,812] [INFO] [timer.py:199:stop] epoch=0/micro_step=1020/global_step=510, RunningAvgSamplesPerSec=27.89634129666697, CurrSamplesPerSec=27.8067088187211, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:11:23,295] [INFO] [logging.py:96:log_dist] [Rank 0] step=520, skipped=5, lr=[0.0007126908421605375, 0.0007126908421605375], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:11:23,299] [INFO] [timer.py:199:stop] epoch=0/micro_step=1040/global_step=520, RunningAvgSamplesPerSec=27.89659063794343, CurrSamplesPerSec=27.815104964319715, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:11:34,808] [INFO] [logging.py:96:log_dist] [Rank 0] step=530, skipped=5, lr=[0.0007026991088541184, 0.0007026991088541184], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:11:34,811] [INFO] [timer.py:199:stop] epoch=0/micro_step=1060/global_step=530, RunningAvgSamplesPerSec=27.895834937739032, CurrSamplesPerSec=27.752305959568453, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:11:46,336] [INFO] [logging.py:96:log_dist] [Rank 0] step=540, skipped=5, lr=[0.0006926095478028312, 0.0006926095478028312], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:11:46,339] [INFO] [timer.py:199:stop] epoch=0/micro_step=1080/global_step=540, RunningAvgSamplesPerSec=27.894509973359995, CurrSamplesPerSec=27.821614456207993, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:11:57,868] [INFO] [logging.py:96:log_dist] [Rank 0] step=550, skipped=5, lr=[0.0006824270284854318, 0.0006824270284854318], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:11:57,871] [INFO] [timer.py:199:stop] epoch=0/micro_step=1100/global_step=550, RunningAvgSamplesPerSec=27.89287517733588, CurrSamplesPerSec=27.847723365469484, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:12:09,396] [INFO] [logging.py:96:log_dist] [Rank 0] step=560, skipped=5, lr=[0.0006721564652446986, 0.0006721564652446986], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:12:09,399] [INFO] [timer.py:199:stop] epoch=0/micro_step=1120/global_step=560, RunningAvgSamplesPerSec=27.891494332148227, CurrSamplesPerSec=27.835600847069383, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:12:20,889] [INFO] [logging.py:96:log_dist] [Rank 0] step=570, skipped=5, lr=[0.0006618028149156478, 0.0006618028149156478], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:12:20,893] [INFO] [timer.py:199:stop] epoch=0/micro_step=1140/global_step=570, RunningAvgSamplesPerSec=27.891413431856982, CurrSamplesPerSec=27.95477589666847, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:12:32,373] [INFO] [logging.py:96:log_dist] [Rank 0] step=580, skipped=5, lr=[0.000651371074433236, 0.000651371074433236], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:12:32,378] [INFO] [timer.py:199:stop] epoch=0/micro_step=1160/global_step=580, RunningAvgSamplesPerSec=27.891824691179327, CurrSamplesPerSec=27.922760285015862, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:12:43,854] [INFO] [logging.py:96:log_dist] [Rank 0] step=590, skipped=5, lr=[0.0006408662784207149, 0.0006408662784207149], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:12:43,858] [INFO] [timer.py:199:stop] epoch=0/micro_step=1180/global_step=590, RunningAvgSamplesPerSec=27.892302295009106, CurrSamplesPerSec=27.9054657757225, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:12:55,346] [INFO] [logging.py:96:log_dist] [Rank 0] step=600, skipped=5, lr=[0.0006302934967597922, 0.0006302934967597922], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:12:55,350] [INFO] [timer.py:199:stop] epoch=0/micro_step=1200/global_step=600, RunningAvgSamplesPerSec=27.8923588635317, CurrSamplesPerSec=27.87588437646747, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:13:03,375] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:13:03,375] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:13:04,497] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 607
[2023-04-14 07:13:04,497] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:13:04,497] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:13:06,800] [INFO] [logging.py:96:log_dist] [Rank 0] step=610, skipped=6, lr=[0.0006207240822732765, 0.0006207240822732765], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:13:06,803] [INFO] [timer.py:199:stop] epoch=0/micro_step=1220/global_step=610, RunningAvgSamplesPerSec=27.894034412979668, CurrSamplesPerSec=27.77138191552118, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:13:18,324] [INFO] [logging.py:96:log_dist] [Rank 0] step=620, skipped=6, lr=[0.0006100362109349642, 0.0006100362109349642], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:13:18,327] [INFO] [timer.py:199:stop] epoch=0/micro_step=1240/global_step=620, RunningAvgSamplesPerSec=27.89279294451914, CurrSamplesPerSec=27.905912527559167, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:13:29,816] [INFO] [logging.py:96:log_dist] [Rank 0] step=630, skipped=6, lr=[0.0005992952333228728, 0.0005992952333228728], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:13:29,820] [INFO] [timer.py:199:stop] epoch=0/micro_step=1260/global_step=630, RunningAvgSamplesPerSec=27.89272171060291, CurrSamplesPerSec=27.96218977944038, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:13:41,310] [INFO] [logging.py:96:log_dist] [Rank 0] step=640, skipped=6, lr=[0.0005885063333059565, 0.0005885063333059565], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:13:41,315] [INFO] [timer.py:199:stop] epoch=0/micro_step=1280/global_step=640, RunningAvgSamplesPerSec=27.89272318651298, CurrSamplesPerSec=27.81198102901513, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:13:52,810] [INFO] [logging.py:96:log_dist] [Rank 0] step=650, skipped=6, lr=[0.0005776747178817414, 0.0005776747178817414], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:13:52,814] [INFO] [timer.py:199:stop] epoch=0/micro_step=1300/global_step=650, RunningAvgSamplesPerSec=27.89249637831583, CurrSamplesPerSec=27.907229658431103, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:14:04,303] [INFO] [logging.py:96:log_dist] [Rank 0] step=660, skipped=6, lr=[0.0005668056146632947, 0.0005668056146632947], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:14:04,307] [INFO] [timer.py:199:stop] epoch=0/micro_step=1320/global_step=660, RunningAvgSamplesPerSec=27.892518411874637, CurrSamplesPerSec=27.875195432893204, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:14:15,778] [INFO] [logging.py:96:log_dist] [Rank 0] step=670, skipped=6, lr=[0.0005559042693562469, 0.0005559042693562469], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:14:15,782] [INFO] [timer.py:199:stop] epoch=0/micro_step=1340/global_step=670, RunningAvgSamplesPerSec=27.893138504741042, CurrSamplesPerSec=27.95382687744928, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:14:27,256] [INFO] [logging.py:96:log_dist] [Rank 0] step=680, skipped=6, lr=[0.0005449759432270804, 0.0005449759432270804], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:14:27,260] [INFO] [timer.py:199:stop] epoch=0/micro_step=1360/global_step=680, RunningAvgSamplesPerSec=27.893536617667916, CurrSamplesPerSec=27.96974165373949, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:14:38,770] [INFO] [logging.py:96:log_dist] [Rank 0] step=690, skipped=6, lr=[0.0005340259105639084, 0.0005340259105639084], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:14:38,773] [INFO] [timer.py:199:stop] epoch=0/micro_step=1380/global_step=690, RunningAvgSamplesPerSec=27.892900012846674, CurrSamplesPerSec=27.69449210326329, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:14:50,296] [INFO] [logging.py:96:log_dist] [Rank 0] step=700, skipped=6, lr=[0.0005230594561309696, 0.0005230594561309696], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:14:50,298] [INFO] [timer.py:199:stop] epoch=0/micro_step=1400/global_step=700, RunningAvgSamplesPerSec=27.891842261813647, CurrSamplesPerSec=27.820172764912243, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:15:00,667] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:15:00,668] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:15:01,790] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 709
[2023-04-14 07:15:01,790] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:15:01,790] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:15:01,791] [INFO] [logging.py:96:log_dist] [Rank 0] step=710, skipped=7, lr=[0.0005131799808136933, 0.0005131799808136933], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:15:01,791] [INFO] [timer.py:199:stop] epoch=0/micro_step=1420/global_step=710, RunningAvgSamplesPerSec=27.891948204255108, CurrSamplesPerSec=28.95740738143756, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:15:13,345] [INFO] [logging.py:96:log_dist] [Rank 0] step=720, skipped=7, lr=[0.000502196910870706, 0.000502196910870706], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:15:13,348] [INFO] [timer.py:199:stop] epoch=0/micro_step=1440/global_step=720, RunningAvgSamplesPerSec=27.890011265201572, CurrSamplesPerSec=27.805216829256672, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:15:24,897] [INFO] [logging.py:96:log_dist] [Rank 0] step=730, skipped=7, lr=[0.000491212780642662, 0.000491212780642662], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:15:24,898] [INFO] [timer.py:199:stop] epoch=0/micro_step=1460/global_step=730, RunningAvgSamplesPerSec=27.888223393268152, CurrSamplesPerSec=27.67360429127508, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:15:36,427] [INFO] [logging.py:96:log_dist] [Rank 0] step=740, skipped=7, lr=[0.00048023289135015165, 0.00048023289135015165], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:15:36,430] [INFO] [timer.py:199:stop] epoch=0/micro_step=1480/global_step=740, RunningAvgSamplesPerSec=27.887041180852446, CurrSamplesPerSec=27.774082919312047, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:15:47,945] [INFO] [logging.py:96:log_dist] [Rank 0] step=750, skipped=7, lr=[0.0004692625421669822, 0.0004692625421669822], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:15:47,948] [INFO] [timer.py:199:stop] epoch=0/micro_step=1500/global_step=750, RunningAvgSamplesPerSec=27.88640588569903, CurrSamplesPerSec=27.785266065711962, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:15:59,438] [INFO] [logging.py:96:log_dist] [Rank 0] step=760, skipped=7, lr=[0.00045830702766266147, 0.00045830702766266147], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:15:59,442] [INFO] [timer.py:199:stop] epoch=0/micro_step=1520/global_step=760, RunningAvgSamplesPerSec=27.886367741109595, CurrSamplesPerSec=27.96386762128507, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:16:10,917] [INFO] [logging.py:96:log_dist] [Rank 0] step=770, skipped=7, lr=[0.0004473716352471042, 0.0004473716352471042], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:16:10,921] [INFO] [timer.py:199:stop] epoch=0/micro_step=1540/global_step=770, RunningAvgSamplesPerSec=27.88680069794766, CurrSamplesPerSec=28.053563105282276, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:16:22,396] [INFO] [logging.py:96:log_dist] [Rank 0] step=780, skipped=7, lr=[0.0004364616426187927, 0.0004364616426187927], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:16:22,398] [INFO] [timer.py:199:stop] epoch=0/micro_step=1560/global_step=780, RunningAvgSamplesPerSec=27.887253810566094, CurrSamplesPerSec=27.89529878735907, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:16:33,902] [INFO] [logging.py:96:log_dist] [Rank 0] step=790, skipped=7, lr=[0.00042558231521762715, 0.00042558231521762715], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:16:33,906] [INFO] [timer.py:199:stop] epoch=0/micro_step=1580/global_step=790, RunningAvgSamplesPerSec=27.886969688438924, CurrSamplesPerSec=27.93280197260283, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:16:45,386] [INFO] [logging.py:96:log_dist] [Rank 0] step=800, skipped=7, lr=[0.0004147389036836881, 0.0004147389036836881], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:16:45,390] [INFO] [timer.py:199:stop] epoch=0/micro_step=1600/global_step=800, RunningAvgSamplesPerSec=27.887305876227373, CurrSamplesPerSec=28.029010161575716, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:16:56,887] [INFO] [logging.py:96:log_dist] [Rank 0] step=810, skipped=7, lr=[0.00040393664132314577, 0.00040393664132314577], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:16:56,891] [INFO] [timer.py:199:stop] epoch=0/micro_step=1620/global_step=810, RunningAvgSamplesPerSec=27.887205733734877, CurrSamplesPerSec=27.867405214544448, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:16:58,024] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:16:58,024] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:16:59,143] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 811
[2023-04-14 07:16:59,143] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:16:59,143] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:17:08,336] [INFO] [logging.py:96:log_dist] [Rank 0] step=820, skipped=8, lr=[0.00039425409710640367, 0.00039425409710640367], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:17:08,340] [INFO] [timer.py:199:stop] epoch=0/micro_step=1640/global_step=820, RunningAvgSamplesPerSec=27.88866067362485, CurrSamplesPerSec=27.88372567949294, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:17:19,826] [INFO] [logging.py:96:log_dist] [Rank 0] step=830, skipped=8, lr=[0.0003835443627787501, 0.0003835443627787501], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:17:19,830] [INFO] [timer.py:199:stop] epoch=0/micro_step=1660/global_step=830, RunningAvgSamplesPerSec=27.888870665784353, CurrSamplesPerSec=27.930802357198182, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:17:31,327] [INFO] [logging.py:96:log_dist] [Rank 0] step=840, skipped=8, lr=[0.00037289083290325663, 0.00037289083290325663], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:17:31,332] [INFO] [timer.py:199:stop] epoch=0/micro_step=1680/global_step=840, RunningAvgSamplesPerSec=27.88868285582777, CurrSamplesPerSec=27.91060138361735, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:17:42,811] [INFO] [logging.py:96:log_dist] [Rank 0] step=850, skipped=8, lr=[0.00036229864914437627, 0.00036229864914437627], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:17:42,815] [INFO] [timer.py:199:stop] epoch=0/micro_step=1700/global_step=850, RunningAvgSamplesPerSec=27.888991103657123, CurrSamplesPerSec=27.99500244663193, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:17:54,311] [INFO] [logging.py:96:log_dist] [Rank 0] step=860, skipped=8, lr=[0.0003517729235593656, 0.0003517729235593656], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:17:54,315] [INFO] [timer.py:199:stop] epoch=0/micro_step=1720/global_step=860, RunningAvgSamplesPerSec=27.888824968158044, CurrSamplesPerSec=27.79561202119477, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:18:05,790] [INFO] [logging.py:96:log_dist] [Rank 0] step=870, skipped=8, lr=[0.0003413187361310768, 0.0003413187361310768], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:18:05,794] [INFO] [timer.py:199:stop] epoch=0/micro_step=1740/global_step=870, RunningAvgSamplesPerSec=27.889214737300335, CurrSamplesPerSec=27.8595326070193, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:18:17,263] [INFO] [logging.py:96:log_dist] [Rank 0] step=880, skipped=8, lr=[0.00033094113231622814, 0.00033094113231622814], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:18:17,267] [INFO] [timer.py:199:stop] epoch=0/micro_step=1760/global_step=880, RunningAvgSamplesPerSec=27.889744389157606, CurrSamplesPerSec=27.939151486387185, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:18:28,755] [INFO] [logging.py:96:log_dist] [Rank 0] step=890, skipped=8, lr=[0.00032064512061033795, 0.00032064512061033795], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:18:28,759] [INFO] [timer.py:199:stop] epoch=0/micro_step=1780/global_step=890, RunningAvgSamplesPerSec=27.889847441807508, CurrSamplesPerSec=27.883441833100377, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:18:40,235] [INFO] [logging.py:96:log_dist] [Rank 0] step=900, skipped=8, lr=[0.0003104356701304984, 0.0003104356701304984], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:18:40,239] [INFO] [timer.py:199:stop] epoch=0/micro_step=1800/global_step=900, RunningAvgSamplesPerSec=27.890308024616438, CurrSamplesPerSec=27.996123612216046, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:18:51,698] [INFO] [logging.py:96:log_dist] [Rank 0] step=910, skipped=8, lr=[0.0003003177082171523, 0.0003003177082171523], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:18:51,703] [INFO] [timer.py:199:stop] epoch=0/micro_step=1820/global_step=910, RunningAvgSamplesPerSec=27.891063811342896, CurrSamplesPerSec=28.02250273037165, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:18:55,133] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:18:55,133] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:18:56,258] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 913
[2023-04-14 07:18:56,258] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:18:56,258] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:19:03,163] [INFO] [logging.py:96:log_dist] [Rank 0] step=920, skipped=9, lr=[0.0002912938021228969, 0.0002912938021228969], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:19:03,167] [INFO] [timer.py:199:stop] epoch=0/micro_step=1840/global_step=920, RunningAvgSamplesPerSec=27.89179147203074, CurrSamplesPerSec=27.909759826857577, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:19:14,663] [INFO] [logging.py:96:log_dist] [Rank 0] step=930, skipped=9, lr=[0.0002813630832692028, 0.0002813630832692028], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:19:14,666] [INFO] [timer.py:199:stop] epoch=0/micro_step=1860/global_step=930, RunningAvgSamplesPerSec=27.891644633622363, CurrSamplesPerSec=27.93588915038034, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:19:26,134] [INFO] [logging.py:96:log_dist] [Rank 0] step=940, skipped=9, lr=[0.0002715378841517797, 0.0002715378841517797], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:19:26,138] [INFO] [timer.py:199:stop] epoch=0/micro_step=1880/global_step=940, RunningAvgSamplesPerSec=27.892197066996538, CurrSamplesPerSec=27.81900798747771, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:19:37,618] [INFO] [logging.py:96:log_dist] [Rank 0] step=950, skipped=9, lr=[0.0002618229466615909, 0.0002618229466615909], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:19:37,623] [INFO] [timer.py:199:stop] epoch=0/micro_step=1900/global_step=950, RunningAvgSamplesPerSec=27.892464622791415, CurrSamplesPerSec=27.991750406629706, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:19:49,136] [INFO] [logging.py:96:log_dist] [Rank 0] step=960, skipped=9, lr=[0.0002522229594745347, 0.0002522229594745347], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:19:49,139] [INFO] [timer.py:199:stop] epoch=0/micro_step=1920/global_step=960, RunningAvgSamplesPerSec=27.89188688703324, CurrSamplesPerSec=27.798715007397167, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:20:00,657] [INFO] [logging.py:96:log_dist] [Rank 0] step=970, skipped=9, lr=[0.00024274255578856863, 0.00024274255578856863], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:20:00,660] [INFO] [timer.py:199:stop] epoch=0/micro_step=1940/global_step=970, RunningAvgSamplesPerSec=27.891210364911945, CurrSamplesPerSec=27.92338186963334, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:20:12,189] [INFO] [logging.py:96:log_dist] [Rank 0] step=980, skipped=9, lr=[0.00023338631108761243, 0.00023338631108761243], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:20:12,191] [INFO] [timer.py:199:stop] epoch=0/micro_step=1960/global_step=980, RunningAvgSamplesPerSec=27.89030240271364, CurrSamplesPerSec=27.735944052707037, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:20:23,713] [INFO] [logging.py:96:log_dist] [Rank 0] step=990, skipped=9, lr=[0.00022415874093330168, 0.00022415874093330168], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:20:23,716] [INFO] [timer.py:199:stop] epoch=0/micro_step=1980/global_step=990, RunningAvgSamplesPerSec=27.88958828343984, CurrSamplesPerSec=27.900581448647156, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:20:35,240] [INFO] [logging.py:96:log_dist] [Rank 0] step=1000, skipped=9, lr=[0.00021506429878566358, 0.00021506429878566358], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:20:35,243] [INFO] [timer.py:199:stop] epoch=0/micro_step=2000/global_step=1000, RunningAvgSamplesPerSec=27.8888042861563, CurrSamplesPerSec=27.843431052527308, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:20:46,760] [INFO] [logging.py:96:log_dist] [Rank 0] step=1010, skipped=9, lr=[0.00020610737385376348, 0.00020610737385376348], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:20:46,763] [INFO] [timer.py:199:stop] epoch=0/micro_step=2020/global_step=1010, RunningAvgSamplesPerSec=27.888221552350714, CurrSamplesPerSec=27.786836451762564, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:20:52,511] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:20:52,511] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:20:53,632] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 1015
[2023-04-14 07:20:53,632] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:20:53,632] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:20:58,235] [INFO] [logging.py:96:log_dist] [Rank 0] step=1020, skipped=10, lr=[0.00019816729286664797, 0.00019816729286664797], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:20:58,237] [INFO] [timer.py:199:stop] epoch=0/micro_step=2040/global_step=1020, RunningAvgSamplesPerSec=27.88870545989438, CurrSamplesPerSec=27.867642445160214, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:21:09,756] [INFO] [logging.py:96:log_dist] [Rank 0] step=1030, skipped=10, lr=[0.00018948350353219912, 0.00018948350353219912], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:21:09,759] [INFO] [timer.py:199:stop] epoch=0/micro_step=2060/global_step=1030, RunningAvgSamplesPerSec=27.88813161815214, CurrSamplesPerSec=27.848382060873252, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:21:21,272] [INFO] [logging.py:96:log_dist] [Rank 0] step=1040, skipped=10, lr=[0.00018094957735583463, 0.00018094957735583463], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:21:21,275] [INFO] [timer.py:199:stop] epoch=0/micro_step=2080/global_step=1040, RunningAvgSamplesPerSec=27.88764375410306, CurrSamplesPerSec=27.903528082688315, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:21:32,763] [INFO] [logging.py:96:log_dist] [Rank 0] step=1050, skipped=10, lr=[0.0001725696330273575, 0.0001725696330273575], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:21:32,767] [INFO] [timer.py:199:stop] epoch=0/micro_step=2100/global_step=1050, RunningAvgSamplesPerSec=27.887679086609435, CurrSamplesPerSec=28.033693636377496, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:21:44,235] [INFO] [logging.py:96:log_dist] [Rank 0] step=1060, skipped=10, lr=[0.00016434771492101485, 0.00016434771492101485], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:21:44,239] [INFO] [timer.py:199:stop] epoch=0/micro_step=2120/global_step=1060, RunningAvgSamplesPerSec=27.88815801794712, CurrSamplesPerSec=28.001240069947958, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:21:55,712] [INFO] [logging.py:96:log_dist] [Rank 0] step=1070, skipped=10, lr=[0.00015628779114358032, 0.00015628779114358032], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:21:55,716] [INFO] [timer.py:199:stop] epoch=0/micro_step=2140/global_step=1070, RunningAvgSamplesPerSec=27.888536944504718, CurrSamplesPerSec=27.99043112291661, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:22:07,186] [INFO] [logging.py:96:log_dist] [Rank 0] step=1080, skipped=10, lr=[0.00014839375161924446, 0.00014839375161924446], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:22:07,190] [INFO] [timer.py:199:stop] epoch=0/micro_step=2160/global_step=1080, RunningAvgSamplesPerSec=27.88894765269856, CurrSamplesPerSec=27.835323752361386, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:22:18,654] [INFO] [logging.py:96:log_dist] [Rank 0] step=1090, skipped=10, lr=[0.0001406694062122389, 0.0001406694062122389], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:22:18,658] [INFO] [timer.py:199:stop] epoch=0/micro_step=2180/global_step=1090, RunningAvgSamplesPerSec=27.889494217031334, CurrSamplesPerSec=28.003711357844125, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:22:30,143] [INFO] [logging.py:96:log_dist] [Rank 0] step=1100, skipped=10, lr=[0.00013311848288809813, 0.00013311848288809813], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:22:30,146] [INFO] [timer.py:199:stop] epoch=0/micro_step=2200/global_step=1100, RunningAvgSamplesPerSec=27.889578327515956, CurrSamplesPerSec=27.854068937808545, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:22:41,668] [INFO] [logging.py:96:log_dist] [Rank 0] step=1110, skipped=10, lr=[0.0001257446259144494, 0.0001257446259144494], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:22:41,671] [INFO] [timer.py:199:stop] epoch=0/micro_step=2220/global_step=1110, RunningAvgSamplesPerSec=27.888904673878535, CurrSamplesPerSec=27.79085813957533, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:22:49,719] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:22:49,719] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:22:50,842] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 1117
[2023-04-14 07:22:50,843] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:22:50,843] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:22:53,446] [INFO] [logging.py:96:log_dist] [Rank 0] step=1120, skipped=11, lr=[0.00011926248951860314, 0.00011926248951860314], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:22:53,450] [INFO] [timer.py:199:stop] epoch=0/micro_step=2240/global_step=1120, RunningAvgSamplesPerSec=27.88276514737834, CurrSamplesPerSec=27.959655920841712, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:23:04,931] [INFO] [logging.py:96:log_dist] [Rank 0] step=1130, skipped=11, lr=[0.00011223479112018653, 0.00011223479112018653], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:23:04,935] [INFO] [timer.py:199:stop] epoch=0/micro_step=2260/global_step=1130, RunningAvgSamplesPerSec=27.882905723669616, CurrSamplesPerSec=27.956068529561303, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:23:16,386] [INFO] [logging.py:96:log_dist] [Rank 0] step=1140, skipped=11, lr=[0.00010539423807301218, 0.00010539423807301218], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:23:16,390] [INFO] [timer.py:199:stop] epoch=0/micro_step=2280/global_step=1140, RunningAvgSamplesPerSec=27.883620594264737, CurrSamplesPerSec=28.00500852036866, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:23:27,875] [INFO] [logging.py:96:log_dist] [Rank 0] step=1150, skipped=11, lr=[9.874413180194608e-05, 9.874413180194608e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:23:27,878] [INFO] [timer.py:199:stop] epoch=0/micro_step=2300/global_step=1150, RunningAvgSamplesPerSec=27.88376230362711, CurrSamplesPerSec=27.839694409482334, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:23:39,400] [INFO] [logging.py:96:log_dist] [Rank 0] step=1160, skipped=11, lr=[9.228768181739628e-05, 9.228768181739628e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:23:39,403] [INFO] [timer.py:199:stop] epoch=0/micro_step=2320/global_step=1160, RunningAvgSamplesPerSec=27.88317799150038, CurrSamplesPerSec=27.911019278934415, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:23:50,930] [INFO] [logging.py:96:log_dist] [Rank 0] step=1170, skipped=11, lr=[8.60280041663225e-05, 8.60280041663225e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:23:50,933] [INFO] [timer.py:199:stop] epoch=0/micro_step=2340/global_step=1170, RunningAvgSamplesPerSec=27.882588082067915, CurrSamplesPerSec=27.84352347062886, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:24:02,445] [INFO] [logging.py:96:log_dist] [Rank 0] step=1180, skipped=11, lr=[7.996811992835184e-05, 7.996811992835184e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:24:02,448] [INFO] [timer.py:199:stop] epoch=0/micro_step=2360/global_step=1180, RunningAvgSamplesPerSec=27.882268201762017, CurrSamplesPerSec=27.892591546585386, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:24:13,955] [INFO] [logging.py:96:log_dist] [Rank 0] step=1190, skipped=11, lr=[7.411095375772925e-05, 7.411095375772925e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:24:13,959] [INFO] [timer.py:199:stop] epoch=0/micro_step=2380/global_step=1190, RunningAvgSamplesPerSec=27.88202009403273, CurrSamplesPerSec=27.86537445431461, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:24:25,476] [INFO] [logging.py:96:log_dist] [Rank 0] step=1200, skipped=11, lr=[6.845933247180514e-05, 6.845933247180514e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:24:25,479] [INFO] [timer.py:199:stop] epoch=0/micro_step=2400/global_step=1200, RunningAvgSamplesPerSec=27.881605191374955, CurrSamplesPerSec=27.84837050454499, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:24:37,000] [INFO] [logging.py:96:log_dist] [Rank 0] step=1210, skipped=11, lr=[6.301598368674105e-05, 6.301598368674105e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:24:37,002] [INFO] [timer.py:199:stop] epoch=0/micro_step=2420/global_step=1210, RunningAvgSamplesPerSec=27.88109042500762, CurrSamplesPerSec=27.851387031683306, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:24:47,327] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:24:47,328] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:24:48,449] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 1219
[2023-04-14 07:24:48,450] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:24:48,450] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:24:48,450] [INFO] [logging.py:96:log_dist] [Rank 0] step=1220, skipped=12, lr=[5.8297216162899295e-05, 5.8297216162899295e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:24:48,451] [INFO] [timer.py:199:stop] epoch=0/micro_step=2440/global_step=1220, RunningAvgSamplesPerSec=27.882041162742162, CurrSamplesPerSec=28.920824294076024, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:24:59,914] [INFO] [logging.py:96:log_dist] [Rank 0] step=1230, skipped=12, lr=[5.325673868567482e-05, 5.325673868567482e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:24:59,918] [INFO] [timer.py:199:stop] epoch=0/micro_step=2460/global_step=1230, RunningAvgSamplesPerSec=27.882582805118812, CurrSamplesPerSec=28.024743707689165, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:25:11,382] [INFO] [logging.py:96:log_dist] [Rank 0] step=1240, skipped=12, lr=[4.843187086769574e-05, 4.843187086769574e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:25:11,386] [INFO] [timer.py:199:stop] epoch=0/micro_step=2480/global_step=1240, RunningAvgSamplesPerSec=27.883115787936894, CurrSamplesPerSec=27.959434594084506, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:25:22,842] [INFO] [logging.py:96:log_dist] [Rank 0] step=1250, skipped=12, lr=[4.38249413128744e-05, 4.38249413128744e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:25:22,846] [INFO] [timer.py:199:stop] epoch=0/micro_step=2500/global_step=1250, RunningAvgSamplesPerSec=27.883773985755955, CurrSamplesPerSec=27.98992912726541, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:25:34,315] [INFO] [logging.py:96:log_dist] [Rank 0] step=1260, skipped=12, lr=[3.9438173442575e-05, 3.9438173442575e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:25:34,319] [INFO] [timer.py:199:stop] epoch=0/micro_step=2520/global_step=1260, RunningAvgSamplesPerSec=27.884217398264887, CurrSamplesPerSec=27.88673248894651, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:25:45,828] [INFO] [logging.py:96:log_dist] [Rank 0] step=1270, skipped=12, lr=[3.5273684422533594e-05, 3.5273684422533594e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:25:45,831] [INFO] [timer.py:199:stop] epoch=0/micro_step=2540/global_step=1270, RunningAvgSamplesPerSec=27.883976240456118, CurrSamplesPerSec=27.828173153796904, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:25:57,331] [INFO] [logging.py:96:log_dist] [Rank 0] step=1280, skipped=12, lr=[3.133348414106035e-05, 3.133348414106035e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:25:57,334] [INFO] [timer.py:199:stop] epoch=0/micro_step=2560/global_step=1280, RunningAvgSamplesPerSec=27.883867373684225, CurrSamplesPerSec=27.819359716918402, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:26:08,841] [INFO] [logging.py:96:log_dist] [Rank 0] step=1290, skipped=12, lr=[2.7619474239016175e-05, 2.7619474239016175e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:26:08,845] [INFO] [timer.py:199:stop] epoch=0/micro_step=2580/global_step=1290, RunningAvgSamplesPerSec=27.883616888716304, CurrSamplesPerSec=27.80841414561024, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:26:20,301] [INFO] [logging.py:96:log_dist] [Rank 0] step=1300, skipped=12, lr=[2.4133447192032476e-05, 2.4133447192032476e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:26:20,305] [INFO] [timer.py:199:stop] epoch=0/micro_step=2600/global_step=1300, RunningAvgSamplesPerSec=27.88425762253477, CurrSamplesPerSec=27.963587966994822, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:26:31,766] [INFO] [logging.py:96:log_dist] [Rank 0] step=1310, skipped=12, lr=[2.087708544541689e-05, 2.087708544541689e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:26:31,770] [INFO] [timer.py:199:stop] epoch=0/micro_step=2620/global_step=1310, RunningAvgSamplesPerSec=27.884794065846652, CurrSamplesPerSec=27.93517398067774, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:26:43,220] [INFO] [logging.py:96:log_dist] [Rank 0] step=1320, skipped=12, lr=[1.7851960602162432e-05, 1.7851960602162432e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:26:43,224] [INFO] [timer.py:199:stop] epoch=0/micro_step=2640/global_step=1320, RunningAvgSamplesPerSec=27.88553456135489, CurrSamplesPerSec=27.976463687615464, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:26:44,359] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:26:44,359] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:26:45,481] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 1321
[2023-04-14 07:26:45,481] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:26:45,481] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:26:54,663] [INFO] [logging.py:96:log_dist] [Rank 0] step=1330, skipped=13, lr=[1.53282648048792e-05, 1.53282648048792e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:26:54,667] [INFO] [timer.py:199:stop] epoch=0/micro_step=2660/global_step=1330, RunningAvgSamplesPerSec=27.886483846959727, CurrSamplesPerSec=27.916034999005806, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:27:06,129] [INFO] [logging.py:96:log_dist] [Rank 0] step=1340, skipped=13, lr=[1.2746419577261248e-05, 1.2746419577261248e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:27:06,133] [INFO] [timer.py:199:stop] epoch=0/micro_step=2680/global_step=1340, RunningAvgSamplesPerSec=27.886975829937917, CurrSamplesPerSec=27.917562132811387, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:27:17,598] [INFO] [logging.py:96:log_dist] [Rank 0] step=1350, skipped=13, lr=[1.0399735319127134e-05, 1.0399735319127134e-05], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:27:17,603] [INFO] [timer.py:199:stop] epoch=0/micro_step=2700/global_step=1350, RunningAvgSamplesPerSec=27.887425274088365, CurrSamplesPerSec=27.92047169795621, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:27:29,066] [INFO] [logging.py:96:log_dist] [Rank 0] step=1360, skipped=13, lr=[8.289344599979375e-06, 8.289344599979375e-06], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:27:29,070] [INFO] [timer.py:199:stop] epoch=0/micro_step=2720/global_step=1360, RunningAvgSamplesPerSec=27.887938002301073, CurrSamplesPerSec=27.977962447777546, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:27:40,552] [INFO] [logging.py:96:log_dist] [Rank 0] step=1370, skipped=13, lr=[6.4162659480493935e-06, 6.4162659480493935e-06], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:27:40,555] [INFO] [timer.py:199:stop] epoch=0/micro_step=2740/global_step=1370, RunningAvgSamplesPerSec=27.88807816395258, CurrSamplesPerSec=27.940076244267264, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:27:52,083] [INFO] [logging.py:96:log_dist] [Rank 0] step=1380, skipped=13, lr=[4.781403358729786e-06, 4.781403358729786e-06], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:27:52,086] [INFO] [timer.py:199:stop] epoch=0/micro_step=2760/global_step=1380, RunningAvgSamplesPerSec=27.887456424438863, CurrSamplesPerSec=27.74971246039417, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:28:03,601] [INFO] [logging.py:96:log_dist] [Rank 0] step=1390, skipped=13, lr=[3.3855458582830455e-06, 3.3855458582830455e-06], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:28:03,604] [INFO] [timer.py:199:stop] epoch=0/micro_step=2780/global_step=1390, RunningAvgSamplesPerSec=27.887111940337512, CurrSamplesPerSec=27.871450259874212, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:28:15,108] [INFO] [logging.py:96:log_dist] [Rank 0] step=1400, skipped=13, lr=[2.2293671230376176e-06, 2.2293671230376176e-06], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:28:15,111] [INFO] [timer.py:199:stop] epoch=0/micro_step=2800/global_step=1400, RunningAvgSamplesPerSec=27.8869534453658, CurrSamplesPerSec=27.841652123641264, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:28:26,626] [INFO] [logging.py:96:log_dist] [Rank 0] step=1410, skipped=13, lr=[1.3134251542544773e-06, 1.3134251542544773e-06], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:28:26,628] [INFO] [timer.py:199:stop] epoch=0/micro_step=2820/global_step=1410, RunningAvgSamplesPerSec=27.886597879837947, CurrSamplesPerSec=27.783661349068428, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:28:38,133] [INFO] [logging.py:96:log_dist] [Rank 0] step=1420, skipped=13, lr=[6.381620088213858e-07, 6.381620088213858e-07], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:28:38,135] [INFO] [timer.py:199:stop] epoch=0/micro_step=2840/global_step=1420, RunningAvgSamplesPerSec=27.88643605360806, CurrSamplesPerSec=27.833776741626135, MemAllocated=8.15GB, MaxMemAllocated=44.21GB
[2023-04-14 07:28:41,567] [INFO] [fused_optimizer.py:370:_update_scale] No Grad overflow for 100 iterations
[2023-04-14 07:28:41,567] [INFO] [fused_optimizer.py:371:_update_scale] Increasing dynamic loss scale from 65536.0 to 131072.0
[2023-04-14 07:28:42,688] [INFO] [fused_optimizer.py:362:_update_scale]
Grad overflow on iteration 1423
[2023-04-14 07:28:42,688] [INFO] [fused_optimizer.py:363:_update_scale] Reducing dynamic loss scale from 131072.0 to 65536.0
[2023-04-14 07:28:42,688] [INFO] [logging.py:96:log_dist] [Rank 0] Overflow detected. Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2023-04-14 07:28:49,374] [INFO] [logging.py:96:log_dist] [Rank 0] step=1430, skipped=14, lr=[2.3647733036713525e-07, 2.3647733036713525e-07], mom=[(0.9, 0.95), (0.9, 0.95)]
[2023-04-14 07:28:49,378] [INFO] [timer.py:199:stop] epoch=0/micro_step=2860/global_step=1430, RunningAvgSamplesPerSec=27.890724382161228, CurrSamplesPerSec=33.86488836407768, MemAllocated=7.87GB, MaxMemAllocated=44.21GB
***** Evaluating perplexity, Epoch 1/1 *****
ppl: 2.1566169261932373
saving the final model ...
[2023-04-14 07:29:39,262] [INFO] [launch.py:460:main] Process 1097 exits successfully.
|