Commits · dee8707e59b85939b11437de11da67e59784957f · OpenDAS / Megatron-LM

".github/workflows/build-wheels-macos.yml" did not exist on "be3f120798183d8e8d8a420f22f2b2223f1b0701"

30 Sep, 2021 3 commits
- Merge branch 'main' into inference · dee8707e
  mshoeybi authored Sep 29, 2021
  
  dee8707e
- added BOS · 24684cbb
  mshoeybi authored Sep 29, 2021
  
  24684cbb
- further refactoring · ff2f0a05
  mshoeybi authored Sep 29, 2021
  
  ff2f0a05
29 Sep, 2021 3 commits
- added multi-batch inference · 390ddef8
  mshoeybi authored Sep 29, 2021
  
  390ddef8
- working · b6555b71
  mshoeybi authored Sep 28, 2021
  
  b6555b71
- working · 6c40f892
  mshoeybi authored Sep 28, 2021
  
  6c40f892
27 Sep, 2021 1 commit
- refactoring, tested and working · 25f9c3f0
  mshoeybi authored Sep 26, 2021
  
  25f9c3f0
24 Sep, 2021 1 commit
- tested and woking · e722c4a9
  mshoeybi authored Sep 23, 2021
  
  e722c4a9
23 Sep, 2021 4 commits
- working · 107c29e8
  mshoeybi authored Sep 22, 2021
  
  107c29e8
- Merge branch 'add_BOS' into 'main' · 5ab64637
  Mohammad Shoeybi authored Sep 22, 2021
```
Add Beginning of Sentence token option and adding semaphore while multi-threading to prevent crashes and hangs due to connection keep-alives

See merge request ADLR/megatron-lm!328
```
  5ab64637
- switching from semaphore to lock · 7b293d9b
  rprenger authored Sep 22, 2021
  
  7b293d9b
- added batching · 2f08c0c3
  mshoeybi authored Sep 22, 2021
  
  2f08c0c3
22 Sep, 2021 3 commits
- sampling · 554d1cc0
  mshoeybi authored Sep 22, 2021
  
  554d1cc0
- sampling · 018c270a
  mshoeybi authored Sep 21, 2021
  
  018c270a
- sampling tested · f1555799
  mshoeybi authored Sep 21, 2021
  
  f1555799
21 Sep, 2021 4 commits
- Merge branch 'fix_ddp_in_tasks' into 'main' · 14f2c684
  Jared Casper authored Sep 21, 2021
```
Fixing memory bug caused by DDP during task

See merge request ADLR/megatron-lm!329
```
  14f2c684
- addressing comments · f65a0f88
  rprenger authored Sep 21, 2021
  
  f65a0f88
- added sampling · 297a5f33
  mshoeybi authored Sep 21, 2021
  
  297a5f33
- Fixing memory bug caused by DDP during task · 9d4fd3d3
  rprenger authored Sep 20, 2021
  
  9d4fd3d3
20 Sep, 2021 5 commits
- Fixing merge conflicts · cb57c380
  rprenger authored Sep 20, 2021
  
  cb57c380
- Merge branch 'inference_context_optimization' into 'main' · 87023abd
  Jared Casper authored Sep 20, 2021
```
Inference context optimization

See merge request ADLR/megatron-lm!321
```
  87023abd
- Inference context optimization · 8b9fe87b
  Mohammad Shoeybi authored Sep 20, 2021
  
  8b9fe87b
- Fixing the URL for the web interface · 7bdeb1e7
  rprenger authored Sep 20, 2021
  
  7bdeb1e7
- Adding the option for beginning of sentence token (and fixing hangs) · 69757f9a
  rprenger authored Sep 20, 2021
  
  69757f9a
17 Sep, 2021 3 commits

Merge branch 'add-temperature-parameter-to-server-api' into 'main' · f47aa770
Jared Casper authored Sep 17, 2021
```
Add temperature to the server API

See merge request ADLR/megatron-lm!325
```
f47aa770

Add temperature to the server API · 527e07c0

Robert Clark authored Sep 10, 2021



A temperature value between 0.0 and 100.0 can now be specified via the API
while running the text generation server. The value passed to
--temperature while running the text generation server is kept as the
default value for all API calls that don't include temperature, even if
different values were manually specified previously.
Signed-Off-By: Robert Clark <roclark@nvidia.com>

527e07c0

Merge branch 'fix_initial_broadcasting' into 'main' · a97d676b
Jared Casper authored Sep 17, 2021
```
Fixes a bug in broadcasting that was causing hanging

See merge request ADLR/megatron-lm!327
```
a97d676b

14 Sep, 2021 5 commits
- Merge branch 'debug_harness' into 'main' · 230633f8
  Ryan Prenger authored Sep 14, 2021
```
Changing API to fix LM Harness Evaluation and make API less confusing/terrible

See merge request ADLR/megatron-lm!326
```
  230633f8
- Fixes a bug in broadcasting that was causing hanging · b46482e8
  rprenger authored Sep 14, 2021
  
  b46482e8
- Addressing comments · 593b47b4
  rprenger authored Sep 14, 2021
  
  593b47b4
- Fixing Merge conflict · 2c500111
  rprenger authored Sep 14, 2021
  
  2c500111
- Changing api to tokens_to_generate, making it so we always generate at least tokens_to_generate · 77979e3b
  rprenger authored Sep 14, 2021
  
  77979e3b
13 Sep, 2021 1 commit
- Changing the interface to the lm eval harness and fixing bugs caused by... · 42e83ee0
  rprenger authored Sep 13, 2021
```
Changing the interface to the lm eval harness and fixing bugs caused by misunderstanding out_seq_length
```
  42e83ee0
10 Sep, 2021 1 commit
- Merge branch 'memory_save' into 'main' · c81bf127
  Jared Casper authored Sep 10, 2021
```
Reducing memory usage during inference

See merge request ADLR/megatron-lm!320
```
  c81bf127
08 Sep, 2021 2 commits
- Merge branch 'remove_generate_samples_gpt' into 'main' · 8fe6f9bf
  Jared Casper authored Sep 08, 2021
```
This file doesn't work withthe new inference code, so it should be remove

See merge request ADLR/megatron-lm!322
```
  8fe6f9bf
- This file doesn't work withthe new inference code, so it should be remove · c6f25b30
  rprenger authored Sep 08, 2021
  
  c6f25b30
07 Sep, 2021 2 commits
- Merge branch 'slym/allreduce-overlap' into 'main' · b462d254
  Jared Casper authored Sep 07, 2021
```
use low-priority stream for nccl overlapping all-reduce and gemm

See merge request ADLR/megatron-lm!319
```
  b462d254
- remove increasing nccl stream for overlapping allreduce and gemm · f3e57f6f
  Sangkug Lym authored Sep 06, 2021
  
  f3e57f6f
03 Sep, 2021 2 commits
- Reducing memory usage during inference · faf58b77
  rprenger authored Sep 03, 2021
  
  faf58b77
- Merge branch 'slym/allreduce_custom' into 'main' · 6e1bde1e
  Jared Casper authored Sep 02, 2021
```
allreduce overlap with wgrad gemm

See merge request ADLR/megatron-lm!316
```
  6e1bde1e