Add prefix prompt generation to GenAI-Perf #208
Merged
Conversation
This file contains 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
nv-hwoo
reviewed
Dec 6, 2024
genai-perf/genai_perf/inputs/retrievers/synthetic_data_retriever.py
Outdated
Show resolved
Hide resolved
genai-perf/genai_perf/inputs/retrievers/synthetic_data_retriever.py
Outdated
Show resolved
Hide resolved
debermudez
previously approved these changes
Dec 6, 2024
nv-hwoo
reviewed
Dec 9, 2024
genai-perf/genai_perf/inputs/retrievers/synthetic_data_retriever.py
Outdated
Show resolved
Hide resolved
genai-perf/genai_perf/inputs/retrievers/synthetic_data_retriever.py
Outdated
Show resolved
Hide resolved
genai-perf/genai_perf/inputs/retrievers/synthetic_data_retriever.py
Outdated
Show resolved
Hide resolved
|
@nv-hwoo Thank you for providing such fast, high-quality reviews. I really appreciate it! I tried to walk through my logic for each of these (one of them is pending expert input). Happy to talk through any of these points on a call too. I'll update the unit tests. |
dyastremsky
force-pushed
the
dyas-system-prompts
branch
from
83e9fc6 to
106851c
Compare
Clarify unit test Remove mentions of system prompt
dyastremsky
force-pushed
the
dyas-system-prompts
branch
from
106851c to
457647a
Compare
dyastremsky
changed the title
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Outdated suggestions cannot be applied.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.
Users can now have GenAI-Perf synthetically generate a system prompt pool by setting --num-system-prompts > 0 and setting a --system-prompt-length (default: 100 tokens). This is useful for benchmarking in caching situations (e.g. with prefix caching)
Screenshot:
Notice that the token count can be off by one, due to concatenating the system prompt and user prompt, creating a final prompt that can have a slight difference in the number of tokens from the two fragments.