Results

Here we have a sortable table of our experiments and the results

---

Performance Star Chart

The chart below shows how each model performed across the most popular top 10 topics by row count. Each line represents a model, and the radial axis represents accuracy.

Make sure you explore what happeened between:

• Base Model -> Final Answer
• Base Model -> Reasoning (Both models) Final Answer
• Base Model -> Final Answer Reasoning (Both models)