¡NUEVO VÍDEO en el LAB! Finalmente la noticia del día de ayer fue la salida del nuevo modelo GROK 2!! Hoy la analizo en el lab, testeo si cumple las expectativas y hablamos de su nuevo matrimonio con Flux 🙂 Link a continuación!
LLMS
-
PrefixLM Architecture and Objective: Non-Causal Training Explained
By
–
My take: There's a prefixlm architecture and a prefixlm objective. – prefixlm arch is just a non casual decoder. – prefixlm objective is a standard causal lm training but with a non casual mask before a random split point (to form inputs/targets) during pretaining. The reason
-
Improve LLM Application Accuracy with New Short Course
By
–
Learn a development pattern to systematically improve the accuracy and reliability of LLM applications in our new short course, Improving Accuracy of LLM Applications, built in partnership with @LaminiAI and @Meta, and taught by Lamini’s CEO @realSharonZhou, and Meta’s Senior… pic.twitter.com/b3YogwxPle
— Andrew Ng (@AndrewYNg) 14 août 2024Learn a development pattern to systematically improve the accuracy and reliability of LLM applications in our new short course, Improving Accuracy of LLM Applications, built in partnership with @LaminiAI and @Meta
, and taught by Lamini’s CEO @realSharonZhou
, and Meta’s Senior -
ChatLLM Playground: Build Interactive Experiences with React, Python, SQL
By
–
Our ChatLLM Playground supports React, allowing you to build interactive experiences, games, and graphs!
— Abacus.AI (@abacusai) 14 août 2024
You can generate and execute Python code, make interactive plots, and search the web. It’s a great way to learn by doing and coding in SQL, python, and React. pic.twitter.com/Yi43azwLgMOur ChatLLM Playground supports React, allowing you to build interactive experiences, games, and graphs! You can generate and execute Python code, make interactive plots, and search the web. It’s a great way to learn by doing and coding in SQL, python, and React.
-
Prompt Caching Reduces Latency by 85% on Long Prompts
By
–
With prompt caching, you can reuse a book's worth of context across multiple API requests. This can also reduce latency by up to 85% on long prompts. Use cases include coding assistants, large document processing, and agentic tool use. Get started: https://
docs.anthropic.com/en/docs/build-
with-claude/prompt-caching
… -
Claude Prompt Caching Reduces Costs Up to 90%
By
–
Prompt caching with Claude. Caching lets you instantly fine-tune model responses with longer and more instructive prompts—all while reducing costs by up to 90%. Available in beta on the Anthropic API today.
-
LongWriter: 10,000+ Word Generation from Long Context LLMs
By
–
New from @thukeg LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs author @realYushiBai is active in discussion section to answer your questions: https://
huggingface.co/papers/2408.07
055
… -
Request for open-weights release of Grok-1.5
By
–
Wow the team is iterating at warp speed, congrats!
Would it be possible to have an open-weights release of Grok-1.5, as you did for 1.0? -
Grok-2 Open-Source Scenario: Speculation on Future AI Strategy
By
–
> Given the trend towards open-source in AI for community engagement, development, and to stick it to the man (or other AI), there's a plausible scenario where Grok-2 could be open-sourced. However, this is speculation based on patterns, not fact. https://
x.com/i/grok/share/N
wQ1BtzXMM5Jv3fVBHhOUelZy
… -
xAI Releases New Model: Full Details on Blog
By
–
Tenéis toda la info del nuevo modelo en el blog de xAI