Asha Anoosheh

Using torch autocast

4

Hi, Can someone confirm if the model is known to work in mixed-precision by simply using torch's autocast function? (or if there's a more proper way intended) Thanks

Refactor Distillation PP support to use new MCore API

2

> [!IMPORTANT] > The `Update branch` button must only be pressed in very rare occassions. > An outdated branch is never blocking the merge of a PR. > Please reach...

Intermediate-tensor distillation support

3

> [!IMPORTANT] > The `Update branch` button must only be pressed in very rare occassions. > An outdated branch is never blocking the merge of a PR. > Please reach...

Implement Speculative transform script for GPT models

1

> [!IMPORTANT] > The `Update branch` button must only be pressed in very rare occassions. > An outdated branch is never blocking the merge of a PR. > Please reach...

> [!IMPORTANT] > The `Update branch` button must only be pressed in very rare occassions. > An outdated branch is never blocking the merge of a PR. > Please reach...

Fix single optional import if ModelOpt not installed

1

> [!IMPORTANT] > The `Update branch` button must only be pressed in very rare occassions. > An outdated branch is never blocking the merge of a PR. > Please reach...

Create separate teacher Layer Spec in KD mode

1

# What does this PR do ? Fixes a bug where models that have specs which depend on number of layers would be problematic between student and teacher models if...

Rename TensorRT Model Optimizer to Model Optimizer

1

# What does this PR do ? Prepare a refactor of the Model Optimizer product name and urls. :warning: For major changes (either in lines of code or in its...

> [!IMPORTANT] > The `Update branch` button must only be pressed in very rare occassions. > An outdated branch is never blocking the merge of a PR. > Please reach...