Running Evals
Metrics
Running Evals
Metrics
You can view and compare metrics across datasets in Athina.
When you run evaluations in Athina, you can view the following metrics:
- Average evaluation score
- Pass Rate: The percentage of examples that pass the evaluation
- Percentile distribution of evaluation scores: The distribution of evaluation scores
You can also compare metrics across datasets side-by-side to understand how your models are performing.