What is this?

This is just a sample prompts from a dataset I collected that aligns over 4 languages. On 3 different settings, it shows how multilinguality is handled in embedding generators.