Go deep on test result history and add multiple criteria to GPT evaluation tests
You can now click on any test to dive deep into the test result history. Select specific date ranges to see the requests from that time period, scrub through the graph to spot patterns over time, and get a full picture of performance.
We’ve also added the ability to add multiple criteria to GPT evaluation tests. Let’s say you’re using an LLM to parse customer support tickets and want to make sure every output contains the correct name, email address and account ID for the customer. You can now set a unique threshold for each of these criteria in one test.
Features
- •UIScrub through the entire history of test results in the individual test pages (See which requests were evaluated per each evaluation window, See results and requests from a specific time period)
- •PlatformImproved GPT evaluation test (Add multiple criteria to a single test, Choose how you want each row to be scored against the criteria: on a range from 0-1, or a binary 0 or 1)
Improvements
- •PerformanceBolster backend server to handle higher loads
- •UITable headers no longer wrap
- •UINull columns hidden in data table
- •UITest metadata moved to the side panel so that the test results graph and data are viewed more easily
- •UISkipped test results are rendered with the most recent result value
- •UITest results graph height increased in the page for an individual test
- •UIDate labels in tests results graph improved
- •PerformanceOnly render rows that were evaluated for GPT metric threshold tests
- •UITest card graphs no longer fla
Fixes
- •UIResults graph was not sorted correctly
- •UITest results graph did not overflow properly
- •UITest results graph did not render all data points
- •MCPEmpty and quasi-constant features test creation was broken
- •UIUndefined column values now rendered as null
- •UIMost recent rows were not being shown by default
- •UILabel chip in results graphs for string validation tests was not inline
- •UITest results were not rendering properly in development mode
- •UIPlan name label overflowed navigation
- •UIButtons for exploring subpopulations was active even when no subpopulations existed
- •UIResults graph rendered loading indicator even after networking completed for skipped evaluations
- •UIRows for the current evaluation window were not rendered in test modals
- •UICommit, metrics, and other tables were not rendering rows
- •UIDuplicate loading and empty placeholders rendered in monitoring mode