Commit Graph

Select branches

Hide Pull Requests

add_xcomposer_testcase

internlm2-reward

main

new_model_release

#1

#10

#100

#106

#111

#114

#116

#120

#121

#123

#124

#125

#126

#128

#132

#133

#136

#138

#139

#14

#140

#141

#142

#143

#144

#145

#147

#151

#152

#153

#154

#155

#156

#158

#159

#160

#161

#163

#165

#166

#170

#174

#175

#176

#178

#180

#182

#183

#184

#185

#188

#189

#190

#192

#193

#194

#195

#196

#197

#198

#199

#200

#201

#203

#204

#206

#207

#208

#210

#211

#212

#213

#214

#216

#217

#218

#219

#220

#222

#223

#224

#225

#226

#227

#228

#229

#23

#230

#231

#233

#238

#239

#24

#240

#242

#243

#245

#246

#247

#249

#25

#250

#251

#252

#254

#255

#256

#257

#259

#260

#261

#262

#263

#264

#265

#266

#27

#271

#272

#273

#274

#275

#276

#277

#279

#281

#282

#284

#285

#286

#289

#292

#293

#294

#295

#297

#298

#299

#30

#300

#301

#302

#303

#304

#306

#307

#308

#309

#310

#311

#312

#313

#314

#315

#317

#319

#32

#320

#321

#322

#324

#325

#326

#328

#329

#330

#338

#339

#34

#342

#345

#352

#354

#359

#36

#360

#362

#367

#373

#374

#375

#376

#378

#379

#38

#380

#382

#383

#39

#391

#392

#393

#396

#397

#398

#4

#40

#403

#404

#405

#407

#408

#411

#412

#413

#417

#418

#419

#420

#421

#422

#423

#424

#426

#427

#428

#429

#43

#436

#439

#44

#440

#443

#448

#449

#450

#451

#452

#453

#454

#455

#456

#460

#462

#464

#465

#466

#467

#468

#470

#471

#474

#475

#476

#477

#478

#479

#48

#480

#481

#484

#485

#486

#488

#489

#490

#491

#492

#493

#495

#496

#498

#499

#5

#502

#504

#506

#507

#51

#510

#512

#513

#514

#515

#516

#517

#518

#519

#52

#520

#521

#522

#523

#524

#526

#527

#528

#529

#530

#531

#532

#533

#534

#536

#537

#538

#539

#540

#541

#542

#543

#544

#545

#546

#547

#548

#549

#550

#551

#554

#555

#557

#560

#562

#563

#564

#565

#566

#567

#569

#57

#570

#577

#580

#582

#583

#583

#585

#586

#587

#588

#589

#59

#590

#591

#593

#594

#595

#596

#597

#598

#599

#6

#601

#605

#61

#611

#613

#616

#619

#620

#621

#622

#624

#625

#626

#627

#629

#63

#632

#633

#634

#635

#636

#637

#638

#640

#641

#65

#650

#651

#652

#658

#663

#666

#667

#67

#674

#689

#692

#695

#7

#703

#705

#710

#717

#72

#721

#727

#73

#730

#732

#735

#740

#744

#745

#746

#751

#752

#753

#754

#755

#764

#767

#769

#77

#78

#780

#786

#795

#80

#802

#803

#805

#806

#814

#816

#817

#818

#819

#820

#821

#822

#824

#827

#835

#838

#89

#9

#91

#97

#98

#99

v0.1.0

v0.2.0

v0.2.1dev20230901

v0.2.1dev20230908

v0.2.1dev20230909

v0.2.1dev20230915

v0.2.1dev20231121

v0.2.1dev20240102

Mono Color

• 1e8f5ebef9 Merge branch 'develop' into feat/init_light_monitoring JiaoPL 2023-11-02 13:31:02 +0800
• 6b2bff421c
  change slurm partition (#464) kkscilife 2023-11-02 13:25:46 +0800
• 354e49f388 change slurm partition wangmengke 2023-11-02 12:27:13 +0800
• 4851291356 fix(optimizer/hybrid_zero_optim.py): fix bucket size full judge condition when reduce scatter overlap huangting4201 2023-11-02 10:30:16 +0800
• ad21459ce4 init light monitoring on all ranks JiaoPL 2023-11-01 22:29:32 +0800
• 3028f07cb7
  fix(readme): update README with original weight download link (#460) Pryest 2023-11-01 14:49:48 +0800
• 0cec3d5607 edit typo and polish Pryest 2023-11-01 14:32:52 +0800
• 83713bacf4 add extra info for original model weights. Pryest 2023-11-01 14:23:33 +0800
• 10b5056e1e fix all-gather overlap the model_checkpoint is 0 yingtongxiong 2023-11-01 12:31:52 +0800
• 21624f6f81
  fix(moe): remove norm&gate force sync (#448) Wenwen Qu 2023-11-01 11:29:55 +0800
• 7aced82ec7 remove some unused function (is norm/gate group) Qu Wenwen 2023-11-01 11:20:16 +0800
• b3def4c162 fix(optimizer/hybrid_zero_optim.py): add reduce_scatter_overlap switch huangting4201 2023-10-31 20:40:58 +0800
• 6b843253eb fix(optimizer/hybrid_zero_optim.py): remove redundant _accum_grad_buckets huangting4201 2023-10-31 20:26:36 +0800
• 4c1cd5d49b fix async reduce scatter mwiacx 2023-10-31 19:39:24 +0800
• 64b89e8e06 update README with original weight download link. Pryest 2023-10-31 15:30:06 +0800
• 2c6cfde332
  fix(web_demo): remove <eoh> in user prompt (#440) vansin 2023-10-27 22:44:30 +0800
• f77f376fd6
  fix(os): fix FileNotFoundError in storage_manager (#455) Yang Gao 2023-10-27 22:32:46 +0800
• 4995060d84
  feat(storage): support ali oss ckpt saving (#439) jiaxingli 2023-10-27 22:32:10 +0800
• bc5a85c624
  Merge pull request #6 from yingtongxiong/fstp/overlap-support-pp ytxiong 2023-10-27 20:32:44 +0800
• 3778c66660 feat(model/overlap_handler.py): fix overlap hander to support pp(non-interleaved) huangting4201 2023-10-27 20:04:23 +0800
• e6d8ebc3e5
  volc_path (#454) jiaxingli 2023-10-27 18:53:06 +0800
• 62e5c2c650 Merge upstream/develop into fix/add_zero_broadcast_sync Qu Wenwen 2023-10-27 17:45:34 +0800
• 87a3c5c374
  feat(optimizer): zero gradient count (#449) jiaopenglong 2023-10-27 16:26:55 +0800
• 4df2c47472 refactor code Qu Wenwen 2023-10-27 15:34:48 +0800
• 739a308c82 fix merged error Qu Wenwen 2023-10-27 15:10:16 +0800
• 950f2de833 Merge upstream/develop into fix/add_zero_broadcast_sync Qu Wenwen 2023-10-27 11:05:53 +0800
• a81a4a4ba8 delete old sync logic Qu Wenwen 2023-10-27 11:05:17 +0800
• c46861f986 fix ci gaoyang07 2023-10-27 10:25:34 +0800
• 2520edb795 use try-except to handle file error gaoyang07 2023-10-27 10:20:43 +0800
• a862f503b6 use rank0 to makedirs gaoyang07 2023-10-26 22:48:38 +0800
• aa3840fc38 fix some bugs yingtongxiong 2023-10-26 20:42:24 +0800
• 87bbef1e0d volc_path li126com 2023-10-26 20:40:56 +0800
• 8aefb74e02 add flash tflops yingtongxiong 2023-10-26 20:33:12 +0800
• 4d83e1021b Merge branch 'feat/fstp_refactor' of https://github.com/yingtongxiong/InternLM into feat/fstp_refactor merge origin yingtongxiong 2023-10-26 20:25:02 +0800
• 3253cbf48e add a new get_tflops_func mwiacx 2023-10-26 20:21:46 +0800
• cbd4f04244 add synchronize yingtongxiong 2023-10-26 20:04:01 +0800
• 325b549707 fix param_metrics is not a tensor JiaoPL 2023-10-26 18:41:13 +0800
• ad70e323eb
  fix(optimizer):broadcast (#453) ytxiong 2023-10-26 17:54:54 +0800
• 42e5f6f8a9
  fix(optimizer):broadcast main (#452) ytxiong 2023-10-26 17:54:48 +0800
• a151615847
  Merge branch 'develop' into fix/broadcast ytxiong 2023-10-26 17:51:48 +0800
• a26481132d fix synchronize yingtongxiong 2023-10-26 17:50:18 +0800
• 4cd803985b
  Merge branch 'main' into fix/broadcast_main ytxiong 2023-10-26 17:48:42 +0800
• c78844626b add synchronize yingtongxiong 2023-10-26 17:46:07 +0800
• 1aae39b667 Merge remote-tracking branch 'upstream/develop' into feat/fstp_refactor merge develop yingtongxiong 2023-10-26 17:41:42 +0800
• d831ddcc1d modify the config yingtongxiong 2023-10-26 17:41:17 +0800
• f653e5af01
  add broadcast synchronize (#451) ytxiong 2023-10-26 17:38:51 +0800
• e4e68ff0ff add broadcast synchronize yingtongxiong 2023-10-26 17:37:08 +0800
• aeee9fd2a9
  fix broadcast synchronize() (#450) ytxiong 2023-10-26 17:33:00 +0800
• cd53d90db9 fix broadcast synchronize() yingtongxiong 2023-10-26 17:29:08 +0800
• 83cb7036a7 add zero_grad_profiling option JiaoPL 2023-10-26 17:20:44 +0800
• 15ff413362 add zero broadcast_sync Qu Wenwen 2023-10-26 16:27:03 +0800
• a6051335b7 fix layer norm with pp JiaoPL 2023-10-26 14:50:54 +0800
• 9ac5ab3101 fix layer norm with pp JiaoPL 2023-10-26 10:07:43 +0800
• e900a1e45f add zero grad count JiaoPL 2023-10-25 18:02:00 +0800
• cc20fa271a reset print memory yingtongxiong 2023-10-25 16:48:02 +0800
• 985465c96a merge upstream yingtongxiong 2023-10-25 14:46:45 +0800
• 363275b500 add memory print yingtongxiong 2023-10-25 14:31:00 +0800
• 1d7e2d04ec
  fix(*)/all-reduce for norm in sequence parallel (#443) ytxiong 2023-10-25 14:16:32 +0800
• 918dff7257 reset moe yingtongxiong 2023-10-25 13:47:19 +0800
• 0bac166b7a add test yingtongxiong 2023-10-25 13:44:15 +0800
• 476a24bd9b fix lint yingtongxiong 2023-10-25 13:38:46 +0800
• 1bc3c33b75 change the order of dp and sp all-reduce yingtongxiong 2023-10-25 13:27:47 +0800
• 1655a90f34 fix all-reduce norm grad yingtongxiong 2023-10-25 13:11:59 +0800
• 41cfa1a10a feat(model/overlap_handler.py): fix overlap handler None bug huangting4201 2023-10-24 18:47:27 +0800
• 0d3592a53f Merge branch 'feat/fstp_refactor' of https://github.com/yingtongxiong/InternLM into feat/fstp_refactor merge origin yingtongxiong 2023-10-24 17:54:50 +0800
• 262de4b796 support tflops computation and generate test py files yingtongxiong 2023-10-24 17:54:26 +0800
• 5d8313693b feat(model/overlap_handler.py): fix head post backward hook when activation huangting4201 2023-10-24 17:29:09 +0800
• 97dcefc389 support model activation checkpoint yingtongxiong 2023-10-24 16:13:52 +0800
• 24d5f05df6
  Update web_demo.py vansin 2023-10-24 07:38:38 +0800
• 12b1001b5e new oss li126com 2023-10-23 20:43:26 +0800
• 7b1b892084
  fix(tools): fix InternLMTokenizer to fit transformers==4.34.0 x54-729 2023-10-23 18:35:30 +0800
• 949a0a1d55
  feat(optimizer): add layer norm to tensorboard (#429) jiaopenglong 2023-10-23 17:07:04 +0800
• 0996c47e49 fix accumulate grads bug chenxun.p 2023-10-23 16:17:57 +0800
• 1e263988cf add function: reduce grads JiaoPL 2023-10-23 15:59:37 +0800
• b48687a7ff
  Merge pull request #5 from yingtongxiong/fstp/refactor-hook-handle huangting4201 2023-10-23 15:35:34 +0800
• b2c1a70477 feat(train/training_internlm.py): fix lint error huangting4201 2023-10-23 15:34:24 +0800
• 9cf1ff0f6e feat(solver/optimizer/hybrid_zero_optim.py): minor update huangting4201 2023-10-23 15:31:41 +0800
• 03cc7f9b80 feat(model/overlap_handler.py): fix lint error huangting4201 2023-10-23 15:28:34 +0800
• 0d693cf3a1 feat(model/overlap_handler.py): fix lint error huangting4201 2023-10-23 15:22:03 +0800
• f6a5086fe4 support bias yingtongxiong 2023-10-23 14:51:27 +0800
• e7f9f1d208 feat(model/overlap_handler.py): optimize reduce scatter mem pool huangting4201 2023-10-23 13:31:23 +0800
• b20f47a1fe feat(model/overlap_handler.py): move handler to gpc huangting4201 2023-10-23 12:02:32 +0800
• 85ad917ae4 feat(model/overlap_handler.py): refactor overlap hook handle huangting4201 2023-10-20 21:50:32 +0800
• 1804d01bb3 merge reduce-scatter yingtongxiong 2023-10-20 18:11:00 +0800
• dcd89ed304 refactor linear yingtongxiong 2023-10-20 17:50:56 +0800
• f22e5b3b28
  Merge pull request #4 from yingtongxiong/fstp/refactor-config ytxiong 2023-10-20 17:48:20 +0800
• 45fd0ec86b test moe layer norm JiaoPL 2023-10-20 16:27:28 +0800
• 2acf9b817f feat(utils/gputest.py): fix lint error huangting4201 2023-10-20 16:25:08 +0800
• eac382ad0a feat(optimizer/hybrid_zero_optim.py): fix lint error huangting4201 2023-10-20 16:22:29 +0800
• 3c6925499f feat(optimizer/hybrid_zero_optim.py): resolve conflicts huangting4201 2023-10-20 16:18:01 +0800
• d91a5d9d9e feat(initialize/launch.py): refactor config for fstp huangting4201 2023-10-20 15:59:40 +0800
• 95488d8e8f update optimizer accumulate grad impl when fstp chenxun.p 2023-10-20 15:58:06 +0800
• 140be20511
  test(workflow): add unit test yaml (#427) kkscilife 2023-10-20 14:22:58 +0800
• 1ac7e4b489 add layer norm to tensorboard JiaoPL 2023-10-20 13:15:59 +0800
• 815a584930 feat(model/linear.py): remove useless code huangting4201 2023-10-20 11:27:59 +0800
• ed7232777a support reduce scatter memory pool yingtongxiong 2023-10-20 10:35:45 +0800
• 68b6b6c454 add main branch changxiaodongTHU 2023-10-19 19:09:01 +0800
• 5084697657 Merge branch 'develop' into ci/unit_test changxiaodongTHU 2023-10-19 18:39:23 +0800
• 3c992a2101
  fix(pipeline): fix interleave type assert and metrics error (#423) Wenwen Qu 2023-10-19 17:29:30 +0800
• b278e28202 add unit test yaml changxiaodongTHU 2023-10-19 17:16:28 +0800