Psych 101 Test: 1,000-10,000 Psychology Evaluation Records

Psych 101 Test is a text-based evaluation suite containing between 1,000 and 10,000 records for benchmarking human cognition models. Created by Marcel Binz in 2024, it serves as the private test set for the 'Centaur' foundation model research.

Use Cases

• Benchmarking foundation models using Psych-101 test items
• Evaluating cognitive reasoning capabilities within the Centaur model framework
• Measuring model accuracy on psychological scenarios provided in the test set

Strengths

• Contains between 1,000 and 10,000 records
• Directly supports the Centaur foundation model paper
• Standardized for the Psych-101 psychology benchmark

Limitations

• Redistribution is strictly prohibited by the author
• CC BY-ND 4.0 license prevents the creation of derivative datasets
• Lack of public documentation for specific column headers

Provenance

Source

Marcel Binz, 'Centaur: a foundation model of human cognition'

Freshness

Last updated October 2024.

Geography

United States