3 ms·
No, it is "do whatever you were already doing with ML, faster" This question seems either from a place of deep confusion or is in bad faith. This post is about
by Cheer2171 3y ago
No, it is "do whatever you were already doing with ML, faster"
This question seems either from a place of deep confusion or is in bad faith. This post is about hardware. The hardware is model independent.* Any issues with models, like hallucinations, are going to be identical if it is run on this platform or a bunch of Nvidia GPUs. Performance in terms of hardware speed and efficiency are orthogonal to performance in terms of model accuracy and hallucinations. Progress on one axis can be made independently to the other.
* Technically no, but close enough
- sorokod 3y agoWell ok, Groq provides lower latency cheaper access to the same models of questionable quality. Is this not putting lipstick on a pig scenario? I suppose more of a question to pig buyers.
- siwakotisaurav 3y agoThey’re probably in the business of being the hardware provider. Best thing would be if Microsoft buys a lot of their chips and that way chatgpt is actually sped up. It’s basically model independent
- Cheer2171 3y agoOkay. How about this: Someone posts to HN about an amazing new battery technology, which they demo by showing an average-sized smartphone watching TikTok endlessly scroll for over 500 hours on a single charge. Then someone comments that TikTok is a garbage fire and a horrible corrupting influence, yadda yadda, all that stuff. They ask: what is the point of making phones last longer just to watch TikTok? They say this improved efficiency in battery tech is just putting lipstick on a pig. That's you in this thread. That's the kind of irrelevant non-contribution you are making here.
- MuffinFlavored 3y agoBatteries are useful. The majority of LLMs are not?
- sorokod 3y agoPerhaps your analogy reveals more then you intended. What does it tell you about the new technology if the best vehicle to demonstrate it is TikTok?
- imtringued 3y agoMixtral 8x7b is competitive with ChatGPT 3.5 Turbo so I'm not sure why you are being so dismissive. https://chat.lmsys.org/ https://chat.lmsys.org/ check the leaderboard.