Your question is RoPE and AliBi Positional Encodings. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
How do positional encodings like RoPE or AliBi work under the hood, and how do they impact the hardware execution of Transformer models on Tenstorrent?