Training vs Model Comparison
Ground Truth Verified
Training Answer
Baseline
-
Accuracy
-
Numerical
-
Completeness
-
Hallucination
VS
-
Model Worse
-
Same
-
Model Better
Model Answer
Under Test
-
Accuracy
-
Numerical
-
Completeness
-
Hallucination
Total Samples
-
-
Accuracy Gap
-
Training vs Model Delta
Numerical Gap
-
Training vs Model Delta
Categories
-
Domains analyzed
Performance by Category
Training vs Model by Category
Accuracy comparisonHallucination Analysis
Training vs ModelDetailed Statistics by Category
| Category | n | Training Acc | Model Acc | Training Num | Model Num | Gap | Status |
|---|
Sample Results
Critical Model Failures
High Hallucination RateQuestions where model answer had significant issues compared to training answer