Post

What do LLMs see? I wrote a lil' tool that extracts the attention matrices out of open models and creates this typing visual, with each token's opacity changing according to its average attention score as the prompt progresses. Dimmer words are considered less important to the model.

So they are narcissists.

Does that imply that beginning a prompt with the most significant topic word might help get more focused results? Or am I missing the point?

My guess is no. The effect of the sink on the first token isn't so much about the first token itself but about moderating attention interactions in other token positions.