I need to think about this more, but I'm proposing token-level calculation of that concept: sum(token_bits / token_bytes for _ in all_tokens) … and not: sum(tokens_bits) / sum(tokens_bytes) Because the second one would hide bad predictions for longer tokens in the sum(),
LLMS
-
AI People releases local LLM for offline NPC gameplay
By
–
As I promised, today we released local LLM for AI People. AI People is the first game ever to feature AI NPCs powered by local LLM – running completely offline on your PC! We still have much work ahead of us to turn AI People into an amazing game! https://
aipeoplegame.com/ai-people-now-
with-local-llm/
… -
Local LLM Released for AI NPCs and New Game Narrative Teaser
By
–
Today was a very productive day: ◆ Released local LLM for AI People – your AI NPCs now run completely offline on your PC ◆ Released a narrative teaser for our new game
-
AI People Releases Local LLM for Offline NPC Gaming
By
–
As I promised, today we released local LLM for AI People. AI People is the first game ever to feature AI NPCs powered by local LLM – running completely offline on your PC! We still have much work ahead of us to turn AI People into an amazing game!
-
Google Gemini adds an experimental chess game
By
–
Google Gemini got updated with a new experimental Chess Gem. Looks like you can play chess with Gemini in a text form. Every reply comes with a respective chessboard image
-
Complex Report Generation with LangGraph Agents
By
–
Complex Report Generation with LangGraph I really like @jxnlco
's take that "RAG" -> "Report Generation" @RLanceMartin explores what this could look like with pretty complex LangGraph agent – give it a watch to start learning about this https://
youtube.com/watch?v=wSxZ7y
Fbbas
… -
Tokenization Boundaries and Cross-Entropy Loss Optimization
By
–
I found the exact tokenization boundary of the text and the corresponding token index. I can do the normalization but it's not perfect… (Testing it shortly.) Computing the cross-entropy like this rewards vocabularies with more tokens and easy predictions (on average) vs.
-
LLM Biorisk Research Lacks Central Question Demonstrations
By
–
There are certain central questions LLM-focused biorisk research needs to demonstrate to us, but we haven’t seen signs of it yet. That, of course, may change.
-
Biorisks LLMs Critical Analysis Open Source Security
By
–
Much of the biorisk work on text-based LLMs struck us as reasonably unconvincing: it often shoed the ability of LLMs to answer a handful of Qs about pathogens…and not much else. @1a3orn
’s takedown of this work, along with their wider writing about attacks on open source, is -
LLM-based agents opportunities and challenges in healthcare medicine
By
–
The opportunities, challenges and outlook for LLM-based agents in medicine and healthcare—our paper published today @NatMachIntell https://
nature.com/articles/s4225
6-024-00944-1
… @jianing_qiu @KyleLam93 @Ara_Darzi @guohao_li @WongTienYin