Folks, game over. I won. GPT is hitting a period of diminishing returns, just like I said it would. x.com/amir/status/18…
LLMS
-
Groq Powers All Submissions at Meta’s Llama Impact Brazil Hackathon
By
–
@AIatMeta
's Llama Impact Brazil, our 1st hackathon in LATAM, was 100% of the submissions were Powered by Groq! The Best on Groq went to Codefire, which strives to reduce barriers to quality education, help students better prepare for the ENEM (Brazilian National High -
GPT-4o Mini vs Llama 3.1: Key Comparison
By
–
GPT-4o Mini vs Llama 3.1. Choose AI to fit your needs. • Assess key features
• Check performance metrics
• Determine ideal use cases Read more: https://
buff.ly/3O7jdhR -
Continued Fine-Tuning Feature Coming Next Week
By
–
👀 Sneak Peek: Continued Fine-Tuning Drops Next Week!
— Predibase by Rubrik (@predibase) 15 novembre 2024
Build on your fine-tuned adapters without starting from scratch. Refine performance or adapt to new datasets—saving time and boosting accuracy.
💡 Want early access? DM us to try it first!#AI #ML #FineTuning #LLMs #SLMs pic.twitter.com/7p6zoTdOA9Sneak Peek: Continued Fine-Tuning Drops Next Week! Build on your fine-tuned adapters without starting from scratch. Refine performance or adapt to new datasets—saving time and boosting accuracy. Want early access? DM us to try it first! #AI #ML #FineTuning #LLMs #SLMs
-
Chinese AI Startup Trains GPT-4 Rival with Minimal Resources
By
–
Chinese AI Startup http://
01.ai Trains GPT-4 Rival with Minimal Resources http://
01.ai, a Chinese AI startup, has trained its advanced model, Yi-Lightning, with just 2,000 GPUs and $3 million, compared to the $80–100 million OpenAI reportedly spent on -
Cerebras Partner Launches TxT360 Trillion Token Dataset
By
–
Cerebras’ partner @mbzuai has announced TxT360 (Trillion eXtracted Text) — the first globally deduplicated dataset across most used data sources for LLM pretraining, and an optimized upsampling recipe to expand to 15T+ tokens of high-quality open-source data for pretraining LLMs.
-
Learning about llms.txt format from Jeremy Howard’s proposal
By
–
Now learning "llms.txt" as an LLM-legible docs format comes from @jeremyphoward who wrote a proposal here: https://
llmstxt.org -
Complete Prompt Engineering Curriculum for AI Agents and RAG Systems
By
–
Complete Prompt Engineering Curriculum Whether you are building an AI agent or complex RAG systems, prompt engineering is key to build effectively. There are lots of approaches out there but in our new LLM prompting courses, we teach you about the methods and best practices
-
Convention for LLM-legible docs at fixed paths like robots.txt
By
–
Sure, but if you're starting a convention for LLM-legible docs (e.g. that docs should always be at some path à la robots.txt) the choice of markdown is incidental; it could be XML, JSON — whatever an LLM understands, which .txt signifies well.
-
OpenAI, Google, Anthropic Struggle Building Next Frontier AI Models
By
–
Happy Friday! If you haven't already, you should read our in-depth piece about how three of the hottest AI companies — OpenAI, Google and Anthropic — are struggling to build their next frontier AI models, and how they're pushing forward. Gift link!