Codonarchitect/genomiratheon_benchmark

GENOMIRATHEON™ Benchmark dataset This repository contains the official GENOMIRATHEON™ prompt-response dataset, designed for evaluating and training language models on synthetic biology compliance simulations. Dataset Overview: genomiratheon_benchmark.json is dataset benchmark consisting of 12 prompt-response pairs focused on codon-level licensing, regulatory hallucination, and LLM compliance alignment. Each entry simulates how future AI systems might respond to… See the full description on the dataset page: https://huggingface.co/datasets/Codonarchitect/genomiratheon_benchmark.