Sketchy’s Post

"LLMs will only be accountable when they can be paired with a contextual knowledge system, specifically a graph. The missing component for attribution for outputs from transformers or LLMs is the component of contextual relays or the contextual backing that can be found in a graph." Listen as our CEO Charlie B. explains the importance of accountability and context in LLMs, and why transparency is key in today's AI landscape: https://lnkd.in/dnnMz4-8 #LLM #KnowledgeGraph #accountabilityinAI

Thinking about how to improve usability and trust of language models these past several years has been a highlight of my intellectual journey.

In case you missed it, Tenyx recently announced our new open-source chat LLM that is top of its class. From the Huggingface page: "At the time of release (Jan 2024), TenyxChat-7B-v1 is the highest-ranked chat model on the MT-Bench evaluation available for download and commercial use." Why is it so good? Because of our innovative fine-tuning technology. You can chat with the model at Huggingface: https://lnkd.in/eskJRWB2 But here is a tutorial for how to deploy the model yourself using Inferless: https://lnkd.in/eeCQgxQQ

I recently worked on this article to explain my perspective on how LLMs are changing multi-agent systems: https://lnkd.in/d7bQxjsZ

Integrate your LLMs with iMerit Ango Hub in a few steps! Easily fine-tune your LLMs and set up your project needs, tools, classification, and workflow. The benefits? ▪️Improved LLM accuracy and performance ▪️High-quality labeled data for fine-tuning ▪️Robust data pipeline with high efficiency Check out this demo video to see the process: https://hubs.ly/Q02tWcT80 #LLM #AngoHub #LLMmodel #datalabeling

Discover how family businesses and private enterprises can safely harness the benefits of GenAI while safeguarding sensitive data. Establish an acceptable use policy for employees to ensure boundaries and guidelines are followed.

If the rise of LLMs caught you by surprise, here's your chance to get a preview of what's likely to be the next monumental jump in AI capabilities: LLM-backed agents that use software tools In this video, I'll walk you through the concepts and code of building an LLM-backed agent with Cohere Command R+ and LangChain.

Discover how family businesses and private enterprises can safely harness the benefits of GenAI while safeguarding sensitive data. Establish an acceptable use policy for employees to ensure boundaries and guidelines are followed. https://pwc.to/3W6xdgo

Yes, if you want to grasp the true significance of LLMs, you must go beyond the conversational workflow. ‘Software is eating the world’, said Marc Andreesen, and AI is no exception to that: even if you’re an LLM, the best way to interact with data and other systems is through code. Allowing AI models to deliberately collect data, process them, and make further decisions based on this, is an extremely potent approach to deliver value beyond the ‘chat with my pdf’ use case. #llm #genai #future

Fantastic explainer for how #agents powered by #largelanguagemodels work and how to implement them in code. #generativeai #machinelearning