Sad that x ai Grok is not getting updated much anymore – no imagen there even
LLMS
-
Small Language Models: The Future of On-Device and Cloud AI
By
–
Exciting Times for Small Models (#SLMs) Check out this new article from our CEO, @devvret_rishi Rishi featured in @SolutionsReview
: https://
solutionsreview.com/the-rise-of-sm
all-language-models-the-future-of-on-device-and-cloud-ai/
… In the article, Dev discusses Apple’s game-changing approach to building GenAI applications with small models and -
SolarPro LLM Now Available on FriendliAI with Impressive Token Generation Speed
By
–
I am thrilled that @upstageai #SolarPro is now on @friendliai
. My testing indicates that the first token takes 0.289 seconds to generate, while each additional token takes 0.004 seconds. This is impressive. Try! https://
x.com/friendliai/sta
/friendliai/status/1837008818863231039
… -
Ultimate Guide to Prompting Techniques and arXiv Navigation
By
–
🆕 The Ultimate Guide to Promptinghttps://t.co/j0wMstDJ2Q
— Latent.Space (@latentspacepod) 20 septembre 2024
with @sanderschulhoff of @LearnPrompting and The Prompt Report!
Timestamps
[00:00:00] Introductions
[00:07:32] Navigating arXiv for paper evaluation
[00:12:23] Taxonomy of prompting techniques
[00:15:46] Zero-shot… pic.twitter.com/x8ZP1UomG7The Ultimate Guide to Prompting https://
latent.space/p/learn-prompt
ing
… with @sanderschulhoff of @LearnPrompting and The Prompt Report! Timestamps
[00:00:00] Introductions
[00:07:32] Navigating arXiv for paper evaluation
[00:12:23] Taxonomy of prompting techniques
[00:15:46] Zero-shot -
Anthropic’s prompt caching enables more efficient LLM algorithms
By
–
Anthropic's prompt caching really should be better known. A lot of features that distinguish LLM vendors are incremental nice-to-haves, but prompt caching enables algorithms otherwise too slow and costly to consider:
-
Google’s Self-Correcting Language Models via Reinforcement Learning
By
–
Hot on Hacker News for AI papers right now Training Language Models to Self-Correct via Reinforcement Learning by Google
-
Stanford paper reveals Chain of Thought unlocks sequential tasks.
By
–
Stanford paper might be the key to OpenAI o1’s performance: What’s so effective about Chain of Thought? ⇒ it unlocks radically different sequential tasks! Reminder: A Chain of Thought (CoT) means that you instruct the model to “think step by step”. Often it’s literally
-
Clarifying tokenization’s role in LLM inference
By
–
I fully concede tokenization causes important problems and an LLM trained without them would be more interesting than this thread. I’m only disproving a specific (but common I think) misunderstanding of its role in inference that does predict against these observations, namely:
-
AI Evolution: o1 Models, Agents, and New Tools Today
By
–
Top stories in AI today: -Sam Altman on AI’s evolution: o1 to agents
-Apple launched its AI Siri update
-Visualize data with AI-powered charts
-Google uses AI to help build cities
-5 new AI tools & 4 new AI jobs Read more: http://
therundown.ai/p/openai-o1-ac
hieves-level-2
… -
New AI Agent Experiments Coming to Google AI Test Kitchen
By
–
"Aigency" and "Backbone" could arrive in the AI Test Kitchen as new experiments