Evaluations metrics dashboard
Metrics from evaluation suite runs are collected in reports that can be viewed in the evaluations metrics dashboard. You can compare aggregate results of Evaluation functions and/or examine the results of individual test cases.
For deeper analysis, viewing LLM traces, or comparisons between runs, select View metrics dashboard in a Logic function view. From there, you can select batches of runs for deep comparisons of metrics, duration, and other benchmarks.
