Hacker News

points by darkwater 4 days ago | 0 comments

  For eg. optimizing skills, it can help to put more important information at the bottom of a skill. 

  LLMs are tuned to overweight very early information so that they don't disregard system prompts. You might be able to exploit this by placing some important facts near the very beginning of a conversation.

Aren't these 2 parts in contradiction then?

[–] sweetdreamerit 3 days ago

https://en.wikipedia.org/wiki/Serial-position_effect

This is also how human memory works

[–] BenoitEssiambre 3 days ago

My impression is that there's only a small area at the beginning of the context window that is artificially boosted, after that DLM becomes important.

Full disclosure: this is info I got from Claude. I was discussing DLM ideas with it a while back and while it mostly confirmed (maybe sycophantically) my ideas, it also said that rather than a pure decaying distribution effect with distance, it was U shaped because of the small boost at the beginning of the context window.