AJRF

10 ms·

181.
by AJRF 2y ago
Hey - thanks for the reply. - Use natural language for telling offloading requirements. Do you mean remove the JSON thing and just summarise the offloading requirements? - Just year of the LLM launch of HF url can help if it’s an
182.
by AJRF 2y ago
I am using Streamlit, and it is the thing that is adding React. I appreciate that it's a heavy site, but just being honest with you - it doesn't seem worth the time optimising this by moving to another lighter framework at this st
183.
by AJRF 2y ago
Jay you seem knowledgeable on this - thanks for answering - I have a question I did look at auto-detecting before, but it seems like you can only really tell the features of a GPU, not so much the good info (VRAM amount and bus speed) - is
184.
by AJRF 2y ago
Not so much purposefully closed source more that I don't want to make it complex by splitting out the data the app uses from the code (co-ordination problem when it comes to deploying that I don't want to deal with for a project o
185.
by AJRF 2y ago
I've now had multiple people ask for this - I will work on adding a new tab for this feature as it is a little different than what the site was originally intended to do. Generally speaking models seem to be bucketed by param count (3b
186.
by AJRF 2y ago
Probably the most niche thing I have ever created, but I recently worked on taking the books and ratings from Patrick Collison's Bookshelf ( https://patrickcollison.com/bookshelf ) and putting them in a spreadsheet with
187.
by AJRF 2y ago
Ah sorry, I will fix that.
188.
by AJRF 2y ago
I recently made a little tool for people interested in running local LLMs to figure out if their hardware is able to run an LLM in GPU memory. https://canirunthisllm.com/
189.
by AJRF 2y ago
It already inadvertently does! The controls Apple has on "who can upload data to Maps" has been gamed for years. Type something like "Double Bedroom" in London - You get Booking.com SEO (MEO?) spam! I had to do some qual
190.
Patrick Collison's 'Bookshelf' in a Spreadsheet(docs.google.com)
2 points by AJRF 2y ago|1 comments
191.
by AJRF 2y ago
I found Patrick's bookshelf article a good source of high value books. I took the data and put it into spreadsheet form so I can do things like add a 'Read' column, add links to buy the books, sort by ratings, etc. Sharing it
192.
by AJRF 2y ago
> But surely quite a lot of the 671 GB or whatever is knowledge that is usually irrelevant? Small correction - It's 671B Parameters - not 671 Gigabytes (doing some rudimentary math if you want to run the entire model in memory it wo
193.
Can I run this LLM? Check if LLM will work on your hardware(canirunthisllm.com)
1 points by AJRF 2y ago|1 comments
194.
by AJRF 2y ago
Hey HN, I've recently started running local LLMs, and one problem I encountered was inconsistent information on whether a model will run within a certain amount of VRAM. I created a simple calculator that helps determine if a model can
195.
by AJRF 2y ago
No I don’t think I agree. There is lots of effort wasted shuffling problems around laterally but not solving for the actual goal, that’s what I am saying.
196.
by AJRF 2y ago
My takeaway is that he didn’t solve anything, he just changed the shape of the problem into one that was familiar to him.
197.
by AJRF 2y ago
I think you’ve missed the point - the rambling is not desired, thinking could be improved if, like another commenter suggested - there was a length penalty applied. You want thinking, but you want to penalise rambling, for many, many reason
198.
by AJRF 2y ago
Just tried hf.co/unsloth/DeepSeek-R1-Distill-Qwen-14B-GGUF:Q4_K_M on Ollama and my oh my are these models chatty. They just ramble on for ages.
199.
by AJRF 2y ago
This seems like extraordinarily weak writing with very little backings or substantial evidence to the claims. That in itself would be fine - if presented as opinion - but this has an air of a historical biography of the actual lineage of th
200.
by AJRF 2y ago
> This is what a good privacy story looks like. Not at all. A good privacy story is not sending this data anywhere.
201.
by AJRF 2y ago
The example he shows here has the answer to his question highlighted in the AI overview, and in huge text in the InfoBox. He got his answer instantly without having to go to another page. Really wish such bold claims had better evidence.
202.
by AJRF 2y ago
>> was never meant to "sell well" Doubly crushing for Apple given it sold worse than their low expectations, they shut the production manufacturing and cut sales expectations from 800k to 400k.
203.
by AJRF 2y ago
Unlikely to change as Gambling companies are major donors to the Labour Party. https://theguardian.com/society/article/2024/jun/28/tory-bet...
204.
by AJRF 2y ago
I presume you are aware of the recent context surrounding this link being posted to hacker news?
205.
by AJRF 2y ago
This comment is quite ironic and funny because the enduring interest in this movie is because it’s heavily disputed that the Final Cut is actually close to being Kubrick intended cut. That’s the whole mythos of the movie and why people like
206.
by AJRF 2y ago
The costume shop owners daughter whispers “you should get a cloak lined with ermine” suggesting she (an underage girl) knows where Bill is going (the mansion party). It feels like there’s at least one scene missing that ties that together a
207.
by AJRF 2y ago
I think they are really overloading that word "Agent". I know there isn't a standard definition - but I think Google are stretching the meaning of that way thinner than most C Suite level execs talk about agents at. I think D
208.
by AJRF 2y ago
Kill Jester
209.
by AJRF 2y ago
Why do you think they are kept in the dark on purpose?
210.
by AJRF 2y ago
> And the logs say, what? You've caught me in my laziness, i've not really learned well how to read logs after a crash. It's inexcusable that I don't know how to do this, and I am now inspired to Google around for how

More ›