181–190 of 197 posts
Earlier quoted context omitted.
It's not an LLM for chat as much as augmentation for search, which a chat like interface, with the LLM used to refine/explain/etc.
How do you know? I'm not seeing any info anywhere.
I also use their lab to test llms like mixtral without having to change the local model in running
I recently downloaded ollama on my Linux machine and even with 3060 12gb gpu and 24 GB Ram I'm unable to run mistral or dolphin and always get an out of memory error. So it's amazing that these companies are able to scale these so well handling thousands of requests per minute. I wish they would do a behind the scenes on how much money, time, optimisation is done to make this all work. Also big fan of anyscale. Their…
So like... what am I supposed to be looking at here? Is it supposed to make me perplexed? > Hello! How can I help you? > I have no idea. I was given this link without any expectation that you could help me. What's this all about? > The concept of "no_search" is a feature that allows users to prevent a search engine from searching the internet for an answer. This feature is being introduced in Bing Chat, as mentioned…
Earlier quoted context omitted.
Just tried and got the same odd response. Maybe "what is this" or is a common phrase searched for that leads to Google Lens? No matter what, Perplexity is now the worst of the worst. They were early with the ability to upload documents but the utter failure of Perplexity to be useful is proving what I have been saying for a year now, (1) LLMs are not "AI" any more than a spell checker is and (2) LLMs are not really u…
Serious question: Why do you think asking it what it is constitutes a good test of its capabilities? Do you think (for example) someone asking you what you are would give them enough ability to judge your abilities? Yes you would get the answer right and the models you are choosing are getting the answer wrong, but you can't extrapolate from that to anything else at all about what this model can or can't do.
It’s a little silly to test a small LLM with a question that at least requires knowledge of its own construction that was not included in its training set, and which really requires existential introspection.
A trick I made up this week is to start a chat with an LLM and only ever say “continue”. I tried this with Bard and at first it says “I’m sorry but I’m going to need more information, what do you want to talk about? Here are some things you could ask: (it provides a list)” but if you just keep prompting it with “continue” then after about 20 cycles the form sort of floats around and it can get kinda weird. It’s real…
go on...
I asked “what is this” and it responded with: Google Lens is an application that allows users to search and identify objects, translate text, and perform various tasks using just a camera or a photo. It is available on both Android and iOS devices. Some key features of Google Lens include: Using the camera to identify objects, such as artwork, plants, or everyday items. Translating text in real-time from over 100 lan…
Google Lens is the first result you get on Google if you search “what is this”. It seems like Google Lens team SEOed their way to the top of Google search and since Perplexity response works by using RAG with search engine content it responds with the info from the top search result plus some own context/hallucination lol.
Earlier quoted context omitted.
Maybe when AI gets good, we can have exciting conversations.
Ah, the day has come when we complain that our artificial humans aren't human enough. What would you do five years ago to get something that can do half the stuff LLMs can do?
If the best screwdriver in the world is advertised as a hammer, expect complaints.
Earlier quoted context omitted.
Ah, the day has come when we complain that our artificial humans aren't human enough. What would you do five years ago to get something that can do half the stuff LLMs can do?
I wouldn't. The day has always been here when we complained about everything. Anyway, I'm not sure you understand the problem. The problem is that there's a chat interface and no instructions. If you try to use it, you'll find it's unsuited for its apparent purpose. If the best screwdriver in the world is advertised as a hammer, expect complaints.
Every thread like this is filled with comments about getting the AI to say something wrong/nonsense. It's incredibly dull conversation.