Model evaluation
Model evaluation ID
Evaluation ID
Recipe configuration
Embedding LLM {{embeddingLLMFriendlyName}}
Chunk Size {{modelEvaluation.embeddingSettings.chunkSizeCharacters == null ? '-' : modelEvaluation.embeddingSettings.chunkSizeCharacters}}
Chunk Overlap {{modelEvaluation.embeddingSettings.chunkOverlapCharacters == null ? '-' : modelEvaluation.embeddingSettings.chunkOverlapCharacters}}
Document Splitting {{modelEvaluation.embeddingSettings.documentSplittingMode == null ? '-' : modelEvaluation.embeddingSettings.documentSplittingMode}}
Completion LLM {{completionLLMFriendlyName}}
Temperature {{modelEvaluation.completionSettings.temperature == null ? '-' : modelEvaluation.completionSettings.temperature}}
Top P {{modelEvaluation.completionSettings.topP == null ? '-' : modelEvaluation.completionSettings.topP}}
Max output tokens {{modelEvaluation.completionSettings.maxOutputTokens == null ? '-' : modelEvaluation.completionSettings.maxOutputTokens}}
Frequency penalty {{modelEvaluation.completionSettings.frequencyPenalty == null ? '-' : modelEvaluation.completionSettings.frequencyPenalty}}
Presence penalty {{modelEvaluation.completionSettings.presencePenalty == null ? '-' : modelEvaluation.completionSettings.presencePenalty}}
Evaluation dataset
Evaluated dataset
Sample configuration
Sample row count
Partitions
  • {{ partition }}
Partition count
Input column
Output column
Context column
Ground truth column
Metrics
{{cur.metricName}} {{cur.formattedValue}}
Evaluation diagnostics
{{message}}
Nothing to report
Metadata
Optional. Informative labels for the model evaluation.
  • Dataset
  • Model
  • Evaluation
  • Custom