Skip to content

Tags: fasterinnerlooper/DeepSpeed

Tags

v0.14.0

Toggle v0.14.0's commit message
Update version.txt

v0.13.5

Toggle v0.13.5's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
fix fused_qkv model accuracy issue (deepspeedai#5217)

Fused_qkv model can not correctly choose the fused_qkv type. Need to
update the module_name_matches.

Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com>

v0.13.4

Toggle v0.13.4's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Add script to check for `--extra-index-url` (deepspeedai#5184)

Co-authored-by: Michael Wyatt <michaelwyatt@microsoft.com>

v0.13.3

Toggle v0.13.3's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Switch cpu-inference workflow from --extra-index-url to --index-url (d…

…eepspeedai#5182)

This switch should cause no impact to the workflow, but ensures that we
only download this package from the correct feed rather than
defaulting/falling back to the index-url
[default](https://pip.pypa.io/en/stable/cli/pip_install/#cmdoption-i)
(PyPI) for if a package existed with a higher version there that would
be chosen instead.

v0.13.2

Toggle v0.13.2's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Remove optimizer step on initialization (deepspeedai#5104)

All ZeRO 1/2/3 stages call the optimizer's `step()` on its
initialization. This increments a counter in the optimizer and produces
a different result in parameter update with the normal usage of PyTorch.
This PR eliminates `step()` in the initialization and lazily configures
some internal states (linking *hp_params*) after the first `step()`
call.

---------

Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com>

v0.13.1

Toggle v0.13.1's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Refactor the Qwen positional emebdding config code (deepspeedai#4955)

follow PR deepspeedai#4920 on Qwen inference code

Co-authored-by: Michael Wyatt <michaelwyatt@microsoft.com>

v0.13.0

Toggle v0.13.0's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Update release.yml

v0.12.6

Toggle v0.12.6's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
Mixtral FastGen Support (deepspeedai#4828)

Adds support for Mixtral with FastGen. Key features implemented:

1. Top-2 MoE support
2. Better support for RoPE thetas
3. The mistral model implementation

---------

Co-authored-by: Michael Wyatt <michaelwyatt@microsoft.com>

v0.12.5

Toggle v0.12.5's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
Fix 4649 (deepspeedai#4650)

Co-authored-by: Michael Wyatt <michaelwyatt@microsoft.com>

v0.12.4

Toggle v0.12.4's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
Add safetensors support (deepspeedai#4659)

Co-authored-by: Logan Adams <114770087+loadams@users.noreply.github.com>
Co-authored-by: Jeff Rasley <jerasley@microsoft.com>
Co-authored-by: Michael Wyatt <michaelwyatt@microsoft.com>