11–20 of 49 posts
Reminds me of Numenta's Thousand Brains Theory of Intelligence.
When this is all settled I suspect we're going to find ourselves the drivers of a chariot, with each horse being an external and artificial mind given direction by our evolved needs.
So a more realistic hope would be: we're the horses, and the drivers are driving us to our carrots.
https://github.com/cg123/mergekit
mergekit is the tool you need to do this https://github.com/cg123/mergekit you can slice off layers and blend models with different strategies.
The dev's blog is great: https://goddard.blog/posts/
...But its not what this paper is describing. They are basically alternating models, AFAIK. Also I have other nitpicks with the paper, like using extremely old/mediocre chat models as bases:
> Pygmillion 6B, Vicuna 13B, Chai Model 6B
I really would like them to compare to Gpt4 instead of claiming victory when matching 3.5. To me GPT4 is the first usable one for a lot of professional uses. 3.5 is fun and gets some stuff right but it’s like a demo.
A Yi 34B or Mixtral finetune on the same data would blow them out of the water. Probably blow ChatGPT 3.5 out of the water as well.
Ok, this seems bunk basically because they never really provide evidence of "better". > ... traditiontal gold-standard approaches use human evaluators that score the quality of generated responses, which can be costly. However, since chat AIs are by definition deployed in social environments with humans, one can leverage statistics of users interaction as a meaningful and aligned measure of chat AI engagingness and q…
Engagement and user retention are directly connected to their bottom line in a way that quality responses (e.g. introducing you to a more fulfilling hobby than chatting with AIs) are not.
The paper refers to ChatGPT as a 175B parameter LLM. This is almost certainly incorrect; the original largest version of GPT-3 was 175B but analysis of the speed and cost of the current model as well as public statements by OpenAI indicate it’s as much as 5-10x smaller.
Ok, this seems bunk basically because they never really provide evidence of "better". > ... traditiontal gold-standard approaches use human evaluators that score the quality of generated responses, which can be costly. However, since chat AIs are by definition deployed in social environments with humans, one can leverage statistics of users interaction as a meaningful and aligned measure of chat AI engagingness and q…
They're making chatbots specifically for humans to waste time with them (a.k.a. entertainment.) Engagement and user retention are directly connected to their bottom line in a way that quality responses (e.g. introducing you to a more fulfilling hobby than chatting with AIs) are not.