This dataset contains 200,000 Sanskrit verb morphology specification prompts focused on tinanta forms (for now). It is designed to train reinforcement learning agents for generating morphologically correct Sanskrit words, grounded in the systematic rules of Pāṇini’s Aṣṭādhyāyī.
