[aas008]

Summary

Add hover tooltips to ISL/OSL profile selectbox options across RHAIIS and LLM-D dashboards. When users open the dropdown and hover over a profile option, a tooltip shows the guidellm workload specification (token counts, samples, turns, etc.).

New files

• custom_dropdown.py — Streamlit's st.selectbox does not support per-option hover tooltips natively — it only allows a single help string for the entire widget. To work around this, custom_dropdown.py injects JavaScript into the parent document via st.components.v1.html(). A MutationObserver watches for [role="option"] elements as the dropdown opens and attaches mouseenter/mouseleave event listeners that show a positioned tooltip with the profile's guidellm specification. Also contains PROFILE_DETAILS (the profile definitions) and tags ISL/OSL help icons with a CSS class for the ℹ️ icon styling.

Changes to existing files

• dashboard.py — Calls inject_profile_tooltips() on page load. Adds help="Hover over dropdown options to see guidellm profile details" to ISL/OSL selectboxes in Performance Plots, Performance Trends, and Compare Versions sections.

• dashboard_styles.py — CSS for the ℹ️ info icon on ISL/OSL selectboxes (scoped via .profile-info-icon class, does not affect other help icons).

• llmd_dashboard.py — Same tooltip integration for LLM-D dashboard: calls inject_profile_tooltips(), restores native st.selectbox (was briefly using a custom HTML dropdown that couldn't return values), adds help= parameter.

Adding tooltips for new profiles

No tooltip appears in the dropdown until an entry is added to PROFILE_DETAILS in custom_dropdown.py:

"4000/512": {
    "name": "Your Profile Name",
    "prompt_tokens": "4000",
    "output_tokens": "512",
},

The k-notation variant (e.g. (4k/512)) is auto-generated. For non-standard display names, add "aliases": ["Custom Name"] to the entry.

Profiles shown in tooltips

Dropdown option	Tooltip
(1k/1k)	Balanced Profile — Input: 1000, Output: 1000
(512/2k)	Heterogeneous — Input: 512 (stdev=128), Output: 2048 (stdev=512)
(2k/128)	Short Prefill-Heavy — Input: 2048, Output: 128
(8k/1k)	Prefill-Heavy — Input: 8000, Output: 1000, Samples: 50
(100k/1k)	Long Context Prefill-Heavy — Input: 100000, Output: 1000, Samples: 10
(8k/800)	Heavy Heterogeneous — Input: 8000 (stdev=8500), Output: 800 (stdev=1500)
Multi-turn	Multi-turn Profile — Input: 128, Output: 128, Turns: 5, Prefix: 512

🤖 Generated with Claude Code

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: 7a782318-2590-4671-b274-d80a5b545826

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}