Human 25K: 25,000 Examples for Biomedical LLM Fine-Tuning

25,000 academic-grade examples for fine-tuning LLMs on human anatomy, physiology, clinical concepts, and biomedical quantitative reasoning. The dataset is structured in a Tiny-Recursive-Model-friendly format and includes a fixed mix of tasks. It was created by WithinUsAI and last updated on January 4, 2026.

Use Cases

• Fine-tuning LLMs for biomedical fact verification based on the 8,000 verification examples
• Training models on self-contained quantitative reasoning based on the 9,000 math examples
• Developing models for clinical concept understanding based on the described task mix

Strengths

• 25,000 examples provide a substantial corpus for training
• Task mix is fixed, offering structured training categories
• Designed in a Tiny-Recursive-Model-friendly format

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is known, but sample data and file formats are unavailable
• License information is unknown, which may restrict usage

Provenance

Source

WithinUsAI

Freshness

Last updated 2026-01-04 02:45:53