Description

This contribution adds the prepared documentation and evaluation materials for the Open Model Evaluation theme.

Included files

• Evaluation workbook containing the prompt bank, scoring rubric and scorecard.
• Research report describing the methodology, model facts and preliminary findings.
• Presentation PDF summarizing the project.
• Presenter notes for explaining the evaluation.

The evaluation compares four open-weight models using correctness, instruction following, clarity and honesty. Recorded scores are preliminary, and further testing and independent review are needed before drawing final conclusions.

Note: These materials are intended for review. Please let me know if they should be placed in a specific submission directory instead.