DeepSeek

Explain scaled dot product attention.

What is RLHF, and why is it used after pretraining?

PPO vs DPO, what's the practical difference?

Fill in the scaled dot product attention formula.

Compare ALiBi's linear distance bias with RoPE's vector rotation. Which one modifies scores additively, and which modifies Q and K directly?

What does ALiBi stand for, and where does it add position in the attention pipeline?

In a 7B transformer, do attention or MLP layers use more FLOPs, and does context length change that?

If attention temperature divides QK^T and sampling temperature divides the output logits, where is each one applied?

What is the goal and risk of injecting noise into pre-softmax attention scores?

Compare top_k and top_p as truncation strategies for sampling