Ask it a question on Turbowarp: https://turbowarp.org/1195458597 Debug: Flag click (Ar output and visual using say) Up arrow (Non Ar output in output_words list) Down arrow (Training) *Left arrow (Reset model) Right arrow (Clear _M _V and grad lists)
Wikipedia: https://en.wikipedia.org/wiki/Transformer_(deep_learning_architecture) *This is a slightly untrained model due to Scratch upload limitations, will release a full version soon #All
