57
Anthropic can now track the bizarre inner workings of a large language model (www.technologyreview.com)
submitted 1 year ago by to c/programming@programming.dev
7 comments fedilink hide all child comments
other discussions
23
Anthropic develops new tool to examine hidden processes in LLM generation (www.technologyreview.com)
submitted 1 year ago* (last edited 1 year ago) by [M] to c/localllama@sh.itjust.works
1 comment fedilink hide all child comments
 

I liked reading this article. Its cool to really poke into the hidden perplexity behind patterns of 'thought' in llms. They aren't merely simple 'auto complete'.

The finding that claude does math in a different way then it says it does and can anticipate words ahead of generation time are facinating.

3
Anthropic can now track the bizarre inner workings of a large language model (www.technologyreview.com)
submitted 1 year ago by [M] to c/technology@lemmy.zip
comment fedilink hide all child comments
 

What they found challenges some basic assumptions about how this technology really works.

1
Anthropic can now track the bizarre inner workings of a large language model (www.technologyreview.com)
submitted 1 year ago by [M] to c/ai_@lemmy.world
comment fedilink hide all child comments