Add vLLM Hook blog post#218
Open
IRENEKO wants to merge 2 commits into
Open
Conversation
…ing vLLM Model Internal States Introduces vLLM Hook, an open-source plugin that exposes vLLM internal states (attention, attention heads, hidden states, activations) for inference-time monitoring and intervention. Covers three use cases — in-model safety guardrails (Attention Tracker), activation steering for instruction following, and selective retrieval for re-ranking — and benchmarks vLLM Hook vs. native vLLM Eagle for hidden-state extraction. Signed-off-by: IRENEKO <cyko@ibm.com>
The folder was relocated under docs/ in IBM/vLLM-Hook to declutter the repo root. Signed-off-by: IRENEKO <cyko@ibm.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Adds a blog post introducing vLLM Hook, an open-source plugin for vLLM that exposes internal states (attention, attention heads, hidden states, activations) for inference-time monitoring and intervention.
The post covers:
ExampleHiddenStatesConnector) for hidden-state extraction across prompt lengths and layer counts