Manually apply 4bit weight packing #7274
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/7274
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit 1be39d1 with merge base de74961 ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D67051119 |
jorgep31415
approved these changes
Dec 10, 2024
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Dec 11, 2024
Summary: ## Context Currently, exporting llama models to Vulkan using 4 bit weight quantization is broken because the behaviour of the `groupwise_affine_quantize_tensor` utility function from `torchao` was recently changed so that the packing of two 4-bit integers into a single 8 bit value does not occur. To fix, just have the `VkInt4WeightOnlyQuantizer` perform that step itself. Reviewed By: jorgep31415 Differential Revision: D67051119
SS-JIA
force-pushed
the
export-D67051119
branch
from
be122f3 to
afaf771
Compare
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Dec 11, 2024
Summary: ## Context Currently, exporting llama models to Vulkan using 4 bit weight quantization is broken because the behaviour of the `groupwise_affine_quantize_tensor` utility function from `torchao` was recently changed so that the packing of two 4-bit integers into a single 8 bit value does not occur. To fix, just have the `VkInt4WeightOnlyQuantizer` perform that step itself. Reviewed By: jorgep31415 Differential Revision: D67051119
SS-JIA
force-pushed
the
export-D67051119
branch
from
afaf771 to
654d8a6
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67051119 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D67051119 |
SS-JIA
added a commit
to SS-JIA/executorch-1
that referenced
this pull request
Dec 12, 2024
Summary: ## Context Currently, exporting llama models to Vulkan using 4 bit weight quantization is broken because the behaviour of the `groupwise_affine_quantize_tensor` utility function from `torchao` was recently changed so that the packing of two 4-bit integers into a single 8 bit value does not occur. To fix, just have the `VkInt4WeightOnlyQuantizer` perform that step itself. Reviewed By: jorgep31415 Differential Revision: D67051119
SS-JIA
force-pushed
the
export-D67051119
branch
from
654d8a6 to
187c22d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67051119 |
Summary: ## Context Currently, exporting llama models to Vulkan using 4 bit weight quantization is broken because the behaviour of the `groupwise_affine_quantize_tensor` utility function from `torchao` was recently changed so that the packing of two 4-bit integers into a single 8 bit value does not occur. To fix, just have the `VkInt4WeightOnlyQuantizer` perform that step itself. Reviewed By: jorgep31415 Differential Revision: D67051119
SS-JIA
force-pushed
the
export-D67051119
branch
from
187c22d to
1be39d1
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67051119 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Context
Currently, exporting llama models to Vulkan using 4 bit weight quantization is broken because the behaviour of the
groupwise_affine_quantize_tensorutility function fromtorchaowas recently changed so that the packing of two 4-bit integers into a single 8 bit value does not occur.To fix, just have the
VkInt4WeightOnlyQuantizerperform that step itself.Differential Revision: D67051119