More here: https://t.co/6GZvHUbHpR
— elvis (@omarsar0) Mar 4, 2024
from Twitter https://twitter.com/omarsar0
March 04, 2024 at 03:45PM
via IFTTT
More here: https://t.co/6GZvHUbHpR
— elvis (@omarsar0) Mar 4, 2024
JUST IN: Anthropic announces Claude 3! Here is what you need to know: • Claude 3 consists of a family of three models (Claude 3 Haiku, Claude 3 Sonnet, and Claude 3 Opus) • Claude 3 Opus (the strongest model) seems to outperform GPT-4 on common benchmarks like MMLU and… https://t.co/QO3FToJHdm https://t.co/t2BNES5i1X
— elvis (@omarsar0) Mar 4, 2024
Wow, Claude 3 is incredible. https://t.co/7qYBWvKf7z
— Matt Shumer (@mattshumer_) Mar 4, 2024
Got it to work. Pass a Pydantic type and a text, receive JSON retrievals from the text. Yeah, DSPy is very cool. https://t.co/4pGMFxng6X https://t.co/kwlG5vOWSR
— Cainã Costa (@cfcosta_) Mar 3, 2024
"You see here I have a sad face ;-( , since tokenization is my least favorite part of working with LLMS, but unfortunately, this is necessary to understand" There are many reasons why people would consider @karpathy as a 🐐 AI Engineer. Tackling essential aspects, even if they… https://t.co/EsphbdtWcc https://t.co/bpSgsrujkG
— Itamar Friedman (@itamar_mar) Mar 3, 2024
I've been generating RPG players by the hundreds thanks to DSPy. And all I need is to define a pydantic class like so: https://t.co/WBsgyONt7T
— Franck SN (@ndzfs) Mar 3, 2024
A lot of good ways to split your data for RAG are dependent on the file type, for instance a Markdown-aware splitter (that can also parse out tables), or a JSON splitter. For pure unstructured text semantic chunking is a good approach. Because you can't make proper use of the… https://t.co/N3aGQZ13iW https://t.co/wGibj6mGQC
— Jerry Liu (@jerryjliu0) Mar 4, 2024