Evaluation prompts were taken from @AnthropicAI hh-rlhf dataset. What’s next? We’ll be releasing more models with higher performance in the future. Stay tuned! P.S,
We're hiring! https://
blog.perplexity.ai/careers
GENERATIVE AI
-
Perplexity releases higher performance AI models, now hiring
By
–
-
Perplexity 70B Outperforms Llama on Refusal Rates
By
–
Our models prioritize intelligence, usefulness, and versatility on an array of tasks, without imposing moral judgments or limitations. Comparing pplx-70b-chat and llama-2-70b-chat (by @AIatMeta ), human evaluators saw: 22.7% fewer “full denials” and 31.9% more “no denials”
-
Perplexity Launches pplx-7b and pplx-70b Chat Models
By
–
We’re excited to announce two models we’ve been training in-house: pplx-7b-chat and pplx-70b-chat! Both these models are built on top of open-source LLMs and fine-tuned for chat. They are available as an alpha release, via Labs and pplx-api. Try it out: https://t.co/2yYQB9jln7 pic.twitter.com/pNxHEWsyjp
— Perplexity (@perplexity_ai) 27 octobre 2023We’re excited to announce two models we’ve been training in-house: pplx-7b-chat and pplx-70b-chat! Both these models are built on top of open-source LLMs and fine-tuned for chat. They are available as an alpha release, via Labs and pplx-api. Try it out: http://
labs.pplx.ai -
Backdrop Build: Free 4-Week Program for Crypto AI Builders
By
–
Looking for support to launch something on Stability? Check out Backdrop Build — a 4 week program for builders in crypto and AI. It's free, async friendly, and there are over $50k in grants for top projects. Apply before October 29 to join @StabilityAI
, @BuildOnBase
, -
NVIDIA Joins MLCommons AI Safety Working Group Initiative
By
–
Generative AI can greatly benefit healthcare and education. NVIDIA is excited to join the @MLCommons AI Safety Working Group to develop AI safety benchmarks and support the global AI ecosystem: https://
nvda.ws/40d2UVV -
Jensen Huang Discusses Generative AI Enterprise Solutions
By
–
Missed @NVIDIA CEO, Jensen Huang, at the #LenovoTechWorld keynote? Click the link to watch Lenovo CEO, Yuanqing Yang, and Jensen Huang discuss how generative AI and hybrid AI solutions will help every enterprise. https://
nvda.ws/46OV3Ae -
Best Day in History for Developers to Build
By
–
Today, Friday October 27th, 2023, is the best day in human history to be a developer. Go build something people love!
-
AI Commercial Productization Milestone and Emerging Social Implications
By
–
I neither feel strongly positively or negatively about this development; mostly mentioning it here as a milestone marker for emerging commercial productization of AI and as an FYI to other people who may suddenly gain surprisingly engaged, articulate fans with infinite capacity.
-
AI-Generated Content Quality Compared to Traditional Blog Spam
By
–
“Would it read to a busy professional as plausibly human?” As someone who has read a lot of blog spam over the years, with production functions that involve varied levels of human activity, this is clearly a better artifact than most blog spam.
-
AI Toolchain Automatically Generates Undergraduate Reviews of Scholar Work
By
–
Expected this but did not expect it to happen so quickly: someone has cobbled together an AI toolchain which scrapes publicly available writing of a target on a topic, produces an ~undergrad level review of it, then publishes, probably to let you butter them up via email.