Bump deepspeed from 0.14.0 to 0.14.1

[dependabot[bot]]

Bumps deepspeed from 0.14.0 to 0.14.1.

Release notes

Sourced from deepspeed's releases.

v0.14.1 Patch release

What's Changed

• Update version.txt after 0.14.0 release by @​mrwyattii in microsoft/DeepSpeed#5238
• Fp6 blog chinese by @​xiaoxiawu-microsoft in microsoft/DeepSpeed#5239
• Add contributed HW support into README by @​delock in microsoft/DeepSpeed#5240
• Set tp world size to 1 in ckpt load, if MPU is not provided by @​samadejacobs in microsoft/DeepSpeed#5243
• Make op builder detection adapt to accelerator change by @​delock in microsoft/DeepSpeed#5206
• Replace HIP_PLATFORM_HCC with HIP_PLATFORM_AMD by @​rraminen in microsoft/DeepSpeed#5264
• Add CI for Habana Labs HPU/Gaudi2 by @​loadams in microsoft/DeepSpeed#5244
• Fix attention mask handling in the Hybrid Engine Bloom flow by @​deepcharm in microsoft/DeepSpeed#5101
• Skip 1Bit Compression and sparsegrad tests for HPU. by @​vshekhawat-hlab in microsoft/DeepSpeed#5270
• Enabled LMCorrectness inference tests on HPU. by @​vshekhawat-hlab in microsoft/DeepSpeed#5271
• Added HPU backend support for torch.compile tests. by @​vshekhawat-hlab in microsoft/DeepSpeed#5269
• Average only valid part of the ipg buffer. by @​BacharL in microsoft/DeepSpeed#5268
• Add HPU accelerator support in unit tests. by @​vshekhawat-hlab in microsoft/DeepSpeed#5162
• Fix loading a universal checkpoint by @​tohtana in microsoft/DeepSpeed#5263
• Add Habana Gaudi2 CI badge to the README by @​loadams in microsoft/DeepSpeed#5286
• Add intel gaudi to contributed HW in README by @​BacharL in microsoft/DeepSpeed#5300
• Fixed Accelerate Link by @​wkaisertexas in microsoft/DeepSpeed#5314
• Enable mixtral 8x7b autotp by @​Yejing-Lai in microsoft/DeepSpeed#5257
• support bf16_optimizer moe expert parallel training and moe EP grad_scale/grad_norm fix by @​inkcherry in microsoft/DeepSpeed#5259
• fix comms dtype by @​mayank31398 in microsoft/DeepSpeed#5297
• Modified regular expression by @​igeni in microsoft/DeepSpeed#5306
• Docs typos fix and grammar suggestions by @​Gr0g0 in microsoft/DeepSpeed#5322
• Added Gaudi2 CI tests. by @​vshekhawat-hlab in microsoft/DeepSpeed#5275
• Improve universal checkpoint by @​tohtana in microsoft/DeepSpeed#5289
• Increase coverage for HPU by @​loadams in microsoft/DeepSpeed#5324
• Add NFS path check for default deepspeed triton cache directory by @​HeyangQin in microsoft/DeepSpeed#5323
• Correct typo in checking on bf16 unit test support by @​loadams in microsoft/DeepSpeed#5317
• Make NFS warning print only once by @​HeyangQin in microsoft/DeepSpeed#5345
• resolve KeyError: 'PDSH_SSH_ARGS_APPEND' by @​Lzhang-hub in microsoft/DeepSpeed#5318
• BF16 optimizer: Clear lp grads after updating hp grads in hook by @​YangQun1 in microsoft/DeepSpeed#5328
• Fix sort of zero checkpoint files by @​tohtana in microsoft/DeepSpeed#5342
• Add distributed_port for deepspeed.initialize by @​LZHgrla in microsoft/DeepSpeed#5260
• [fix] fix typo s/simultanenously /simultaneously by @​digger-yu in microsoft/DeepSpeed#5359
• Update container version for Gaudi2 CI by @​raza-sikander in microsoft/DeepSpeed#5360
• compute global norm on device by @​BacharL in microsoft/DeepSpeed#5125
• logger update with torch master changes by @​rogerxfeng8 in microsoft/DeepSpeed#5346
• Ensure capacity does not exceed number of tokens by @​jeffra in microsoft/DeepSpeed#5353
• Update workflows that use cu116 to cu117 by @​loadams in microsoft/DeepSpeed#5361
• FP [6,8,12] quantizer op by @​jeffra in microsoft/DeepSpeed#5336
• CPU SHM based inference_all_reduce improve by @​delock in microsoft/DeepSpeed#5320
• Auto convert moe param groups by @​jeffra in microsoft/DeepSpeed#5354
• Support MoE for pipeline models by @​mosheisland in microsoft/DeepSpeed#5338
• Update pytest and transformers with fixes for pytest>= 8.0.0 by @​loadams in microsoft/DeepSpeed#5164
• Increase CI coverage for Gaudi2 accelerator. by @​vshekhawat-hlab in microsoft/DeepSpeed#5358
• Add CI for Intel XPU/Max1100 by @​Liangliang-Ma in microsoft/DeepSpeed#5376
• Update path name on xpu-max1100.yml, add badge in README by @​loadams in microsoft/DeepSpeed#5386
• Update checkout action on workflows on ubuntu 20.04 by @​loadams in microsoft/DeepSpeed#5387
• Cleanup required_torch_version code and references. by @​loadams in microsoft/DeepSpeed#5370

... (truncated)

Commits

• e3d873a Fix the FP6 kernels compilation problem on non-Ampere GPUs. (#5333)
• 54c0687 stage3: efficient compute of scaled_global_grad_norm (#5256)
• 7b5b066 fix pagable h2d memcpy (#5301)
• f69f884 Removal of cuda hardcoded string with get_device function (#5351)
• 2c51aba Add custom reshaping for universal checkpoint (#5390)
• a8b8215 Optimize the fp-dequantizer to get high memory-BW utilization (#5373)
• cc9e7b9 Update README.md for intel XPU support (#5389)
• 6dcced1 Cleanup required_torch_version code and references. (#5370)
• 63029e8 Update checkout action on workflows on ubuntu 20.04 (#5387)
• 62ab181 Revert "Update XPU workflow too"
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot show ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[dependabot[bot]]

Superseded by #55.