The two tricks I'm trying with Haiku that seem to work incredibly well are feeding it a small targeted sequence of fake "user"/"assistant" examples in the messages array and finishing with a prefilled "assistant" message to point it in the right direction
GENERATIVE AI
-
RAG and LLM: Advancing Dynamic Language Modeling Frontier
By
–
RAG and LLM: A New Frontier in Dynamic Language Modeling
by @_odsc Read more: https://
buff.ly/4129LBL #AI #BigData #DataScience #MachineLearning #DeepLearning cc: @pascal_bornet @pbalakrishnarao @rtehrani -
Metaforms AI: OpenAI meets Typeform innovation
By
–
Yes, Typeform is shit expensive, but that's because they've not upgraded at all.
— Akshat Metaforms (@ofcAkshat) 4 avril 2024
It's exactly the same single-scree interface that it was 10 years ago.
Imagine if OpenAI and Typeform had a kid— Introducing Metaforms AI pic.twitter.com/wa2D4eWEElYes, Typeform is shit expensive, but that's because they've not upgraded at all. It's exactly the same single-scree interface that it was 10 years ago. Imagine if OpenAI and Typeform had a kid— Introducing Metaforms AI
→ View original post on X — @sudalairajkumar, 2024-04-04 09:07 UTC
-
Assessing LLM Brain Damage: Rapid Evaluation Methods
By
–
how do you assess brain damage of an llm this quickly?
-
3Blue1Brown’s Visual Guide to GPT and Transformers
By
–
In case you missed it, 3Blue1Brown has released a video on "But what is a GPT?Visual intro to Transformers" couple of days back. Even if you have a good knowledge on Transformers, this is highly recommended. Visual illustrations help us grasp the underlying concepts easily. Link: piped.video/watch?v=wjZofJX0… Looking forward to the next two videos of the series. Happy learning!
→ View original post on X — @sudalairajkumar, 2024-04-04 04:51 UTC
-
Suno is Just the Beginning of 2024 AI Music Revolution
By
–
u would not believe what i just previewed today. suno is just the start of this year in ai music
-
Open weights embedding models gaining traction over closed APIs
By
–
I find it hard to get excited about closed API-only embedding models these days, because open weights ones (eg Nomic 1.5) are getting REALLY good I'd rather use on open weights model available via an API, safe in the knowledge that if the API ever goes away I can run it myself
-
Statistical Literacy Requires Priors and Baselines for JSON Mode
By
–
you need priors and baselines to be statistically literate. i expected json mode to be <half of fn calling, esp since fn calling is more than twice as old.
-
Groq Launches Tool Calling with LangChain Structured Output Support
By
–
Groq tool calling + structured output @GroqInc just dropped tool calling! We've added LangChain support (including the popular `withStructuredOutput` method!) so you can try it in your favorite chains and apps. It supports @MistralAI Mixtral, Llama 70B, and Google
-
Bay Area AI Hackathon: Build and Monetize Bots
By
–
This hackathon is in partnership with @cerebral_valley and @SHACK15sf
, and will feature a keynote from @adamdangelo and includes hands-on support from the Poe team. If you’re in the Bay Area, come build a bot and monetize it: https://
partiful.com/e/5e04bH25WI0V
kx7XwZMj
… (4/4)