Medical reasoning records targeting MedXpertQA and MedHELM. Each entry includes a mechanistic explanation.
Available on Request · Sample Available
Scalable Volume
Medical Reasoning, Healthcare
MedXpertQA (Text), HealthBench Hard, MedHELM
Medical reasoning records targeting MedXpertQA and MedHELM. Each entry includes a mechanistic explanation.
| Field | Description |
|---|---|
answer | The concise final answer associated with the question. |
question | The question or task presented in the dataset record. |
solution | The worked solution or supporting reasoning for the question. |
| Model | Metric | Result |
|---|---|---|
| Opus 5 | Pass@1 | 65% |
| Opus 5 | Pass@3 | 70% |