Skip to content

Conversation

@ankrgyl
Copy link
Contributor

@ankrgyl ankrgyl commented Mar 1, 2025

At the time we implemented this, LLMs were both more expensive and more wordy. Now, 512 is not always an appropriate limit, so let the LLM pick (by default).

@ankrgyl ankrgyl requested a review from ibolmo March 1, 2025 16:49
@github-actions
Copy link

github-actions bot commented Mar 1, 2025

Braintrust eval report

Autoevals (remove-max-tokens-1740847882)

Score Average Improvements Regressions
NumericDiff 75.2% (+0pp) 1 🟢 -
Start 1740847882.15s - -
End 1740847885.36s - -
Duration 5.18s (+3.62s) - 104 🔴
Llm_duration 1.51s (-0.02s) 3 🟢 1 🔴
Prompt_tokens 279.24tok (+0.02tok) - 3 🔴
Completion_tokens 20.63tok (+0tok) - -
Total_tokens 299.87tok (+0.02tok) - 3 🔴
Estimated_cost 0$ (+0$) - -

Copy link
Collaborator

@ibolmo ibolmo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants