tikkun

35 ms·

421.
by tikkun 3y ago
Can someone else confirm the url works? It may require you to be signed into twitter to see it.
422.
Excuse me, is there a problem?(longform.asmartbear.com)
3 points by tikkun 3y ago|0 comments
423.
by tikkun 3y ago
Clickable: https://twitter.com/vercel/jobs
424.
Twitter now lists job postings for some companies
4 points by tikkun 3y ago|6 comments
425.
AI Playground(sdk.vercel.ai)
4 points by tikkun 3y ago|0 comments
426.
Llama 2 Resources(github.com)
2 points by tikkun 3y ago|0 comments
427.
by tikkun 3y ago
Slight update in the linked post, adding <s> at the beginning
428.
by tikkun 3y ago
My guess is I’ll probably update it while things change quickly, and then once stuff settles down leave it as is
429.
by tikkun 3y ago
Hosted APIs: 70B chat: https://replicate.com/replicate/llama70b-v2-chat 13B chat: https://replicate.com/a16z-infra/llama13b-v2-chat Or run it on a cloud GPU and self-host an API. I've posted
430.
Llama 2 Resources(gpus.llm-utils.org)
3 points by tikkun 3y ago|3 comments
431.
by tikkun 3y ago
What else should I add?
432.
by tikkun 3y ago
The prompt template on the quantized versions of Llama 2 appears to be incorrect relative to the official Meta one ( https://github.com/facebookresearch/llama/blob/main/llama/ge... ). I took Meta&#x27
433.
Llama 2 Prompt Template(gpus.llm-utils.org)
2 points by tikkun 3y ago|2 comments
434.
Superhuman AI(blog.superhuman.com)
2 points by tikkun 3y ago|0 comments
435.
Running Llama 2 on Runpod with Oobabooga's text-generation-webui(gpus.llm-utils.org)
3 points by tikkun 3y ago|0 comments
436.
by tikkun 3y ago
You can use Just Press Record which is good as the other commenter mentioned. The other option is to use the Voice Memos app on Watch and then use the voice memos app on iPhone to share it to something like Otter.ai. If you want to bulk upl
437.
by tikkun 3y ago
FWIW, I emailed him at the same time I posted the comment :)
438.
Ask HN: Have you run into software issues with H100 GPUs?
3 points by tikkun 3y ago|1 comments
439.
by tikkun 3y ago
Thanks
440.
by tikkun 3y ago
> An error occurred > Sorry, the page you are looking for is currently unavailable. > We apologise for the inconvenience, please try again later. And it's back: Household saving has whipsawed over the past three years. The adj
441.
by tikkun 3y ago
@dang, may be good to change the title to 'in Norway'
442.
GPTQ vs. GGML vs. Base Models: Speed and VRAM Test for Vicuna-33B on A100 GPU(gpus.llm-utils.org)
1 points by tikkun 3y ago|0 comments
443.
by tikkun 3y ago
I hope it’s good! What’s the main difference relative to the other tools that work on this? I’ve used probably a dozen movie finding tools over the years including a few that I paid for and liked quite a few of them. Currently mostly use go
444.
by tikkun 3y ago
What does active threads represent?
445.
Ways LLMs could be used in existing products to add a tiny bit of convenience(gpus.llm-utils.org)
12 points by tikkun 3y ago|4 comments
446.
by tikkun 3y ago
What about for you: What are some regular occurrences where you notice that if that product had an LLM integrated, it'd ever so slightly improve the product experience? (I think this is a decent source of startup ideas, too)
447.
by tikkun 3y ago
Could you help me understand? I see that they don't offer what you're seeking, which means I don't understand what you're seeking. I know they allow you to see different variations of what items and where those items are
448.
AI Products That Small Businesses Want(gpus.llm-utils.org)
1 points by tikkun 3y ago|0 comments
449.
by tikkun 3y ago
> GitHub PRs/Issues where devs implement and discuss the paper are really awesome. Could you link two or three examples of this that you've seen and liked?
450.
by tikkun 3y ago
Yep lots: airoomplanner.com interiorai.com reimaginehome.ai dreamhouseai.com roomai.com

More ›