Documentation Index
Fetch the complete documentation index at: https://docs.latitude.so/llms.txt
Use this file to discover all available pages before exploring further.
Score Analytics
Score analytics show quality trends across your project: whether quality is improving, which evaluations catch the most failures, and when issues occur.Project-Level Dashboard
The project overview shows:- Pass/fail distribution: How many scores passed or failed over time
- Failure rate trend: The percentage of failing scores over days or weeks
- Score volume: Total scores, broken down by source
Evaluation-Level Analytics
Each evaluation has its own analytics page with:- Pass/fail trend: How results change over time
- Value distribution: A histogram of score values
- Volume: How many traces the evaluation has scored
- Alignment: Whether the evaluation agrees with human review when annotations exist for the same traces
Issue-Level Analytics
Each issue tracks:- Occurrence count: How many times the issue has been detected
- Lifecycle state: Whether the issue is new, escalating, resolved, or regressed
- Resolution history: When it was resolved and whether it has returned
Score-Aware Trace Filtering
Traces and sessions can be filtered by score-derived properties:- Score state: Failing scores, passing scores, or draft annotations
- Value thresholds: Scores below a quality threshold
- Issue linkage: Traces associated with a specific issue
- Score source: A specific evaluation, annotation source, or custom source
Filtering Analytics
Analytics dashboards use the same filter system as trace views. Narrow analytics by time range, model, provider, score source, or custom metadata to answer targeted questions such as: “What is the failure rate for GPT-4 traces in production this week?”Next Steps
- Scores Overview: How the score model works
- Evaluations: How automated evaluations produce scores
- Issues: How failure patterns are discovered from scores