$\dot x = \sigma(y-x)$, $\dot y = x(\rho - z) - y$, $\dot z = xy - \beta z$ with $\sigma = 10$ and $\beta = 8/3$. The largest Lyapunov exponent $\lambda_1$ comes from one trajectory per $\rho$ after a warm-up, through DynamicalModels.lyapunov_exponent. Where $\lambda_1 > 0$ nearby trajectories separate: the system is chaotic. The textbook onset for these $\sigma, \beta$ is near $\rho \approx 24.74$.