The future happened when a student asking for a letter of recommendation sent me the suggested GPT-4 prompt he wanted me to use to generate the letter itself. Likely where recommendation letters are eventually going to go: 1-10 scale & a single sentence assessment. No ceremony.
LLMS
-
Evaluation of AI search tool indexing and data freshness
By
–
The disappointing part is that it relies on some sort of index like other AI competitors and doesn't pull the most up-to-date info (like the recent profile description for instance). But the good part is that this index is still quite fresh
-
Long-Context Windows vs RAG: Finding the Middle Ground
By
–
And then there's the emergence of long-context windows and whether *they* will also offer an alternative to RAG. But the reality is, once again, probably somewhere in the middle! Here's what @brittwalker_ told me.
-
Vector Databases Evolution: One Year After ChatGPT
By
–
One of my first issues for Supervised (almost a year ago!!) was covering the hype around vector databases following the release of ChatGPT and, perhaps more importantly, the leak of Llama 1. I felt this was as good a time as any to check back in on how things have developed.
-
LLM Bias in HR: Papers Reveal Hidden Discrimination Risks
By
–
This is a useful thread of recent papers showing that LLMs are biased when judging people in HR contexts. The papers offer some potential solutions, but it is important to recognize that current systems carry subtle biases that can be hard to recognize without extensive testing.
-
Programming LLMs with DSPy Instead of Prompting
By
–
Program your LLMs not prompting. Nice thread Akshay on DSPy!
-
Command R and R+ Models Now Available on AWS Marketplace
By
–
Command R (H100) https://
aws.amazon.com/marketplace/pp
/prodview-jfhyfeewxqbr2
… Command R (A100) https://
aws.amazon.com/marketplace/pp
/prodview-w7ukdez7zfjfo
… Command R+ (H100) https://
aws.amazon.com/marketplace/pp
/prodview-jnw3ilzmy7tcg
… Command R+ (A100) https://
aws.amazon.com/marketplace/pp
/prodview-xfznx42jopjfy
… -
Cohere Command R Models Now Available on AWS SageMaker
By
–
More exciting news, our latest Command R and R+ models are now accessible via @awscloud SageMaker! This further builds on our collaboration and allows enterprises to start deploying our RAG-optimized models at production scale. Try our new models on SageMaker today.
-
Claude 3 Opus and Sonnet 200k Models Now Available
By
–
We've added Claude 3 Opus 200k and Sonnet 200k as bases for your prompt bots! These multimodal models are the most powerful base bots available, and can handle longer inputs for more natural back-and-forth conversations, in-depth analysis, multi-step tasks, and more. pic.twitter.com/QKjGlkCTjY
— Poe (@poe_platform) 5 avril 2024We've added Claude 3 Opus 200k and Sonnet 200k as bases for your prompt bots! These multimodal models are the most powerful base bots available, and can handle longer inputs for more natural back-and-forth conversations, in-depth analysis, multi-step tasks, and more.
-
Downloading AI Models to External Storage Directories
By
–
What's the recipe for telling it to download the model to a specific directory? I need to run that on an external volume