Claude Opus breaks down the task into subtasks, then deploys instances of Haiku to solve each task. Once these tasks are solved, the output goes back to Opus for review. Pretty incredible. When I tried to build things like this with GPT4, it was very hard for the model to know
GENERATIVE AI
-
Claude’s Full Output Thinking Process Saved and Analyzed
By
–
This is the full output saved as a text file, it's fascinating to see how Claude thinks. The whole process took exactly 13 minutes and 20 seconds.
-
DHS AI pilot programs raise civil liberties concerns
By
–
Today the US DHS announced three "pilot" use cases of "AI" (by which they apparently mean synthetic text extruding machines). Despite their repeated assurances of "rigorous testing" and "protecting civil liberties" I am … just a bit (a lot) skeptical:
-
X Risk Community Mobilizes Against AI Deepfakes and Job Displacement
By
–
“X risk community” has adapted to this by telling the normies to sign petitions against AI doing deepfake porn and making everybody unemployed
-
Frontier Models, RAG, and AI Agents in Enterprise
By
–
In addition to reviewing the positives & negatives of the three frontier models, this post has some of my more provocative takes for people using AI inside organizations, including why RAG may not be the future, and why agents like Devin the Coder might be
-
TheProfessor: Advanced LLM Merging for Conversational AI and Scientific Applications
By
–
TheProfessor is a merge of #LLMs, created with help from @AbacusAI
's #GenerativeAI team. It can be used for many things with its conversational #AI, reasoning, scientific, medical, & mathematical skills interactive brainstorming: https://
huggingface.co/abacusai/ThePr
ofessor-155b
… -
Local Bot Quest for Automated Writing Tasks
By
–
tbf writing is my least favorite part of this job but my year-long quest to build a local bot (explicitly not an API) that can successfully do this part of my job for me has thus far not been successful
-
DeepLearningAI Course on Efficiently Serving LLMs Launched
By
–
Excited to share the launch of the @DeepLearningAI course Efficiently Serving #LLMs developed w/ @predibase CTO @TravisAddair Implement a modern inference stack w/ caching, batching & #quantization Intro to #LoRA adapters
…and much more -
RAG Query Routing: Handling Multiple Datastores
By
–
RAG From Scratch: Routing Our RAG From Scratch video series walks through impt RAG concepts in short / focused videos w/ code. This is the 10th video in our series and focuses on query routing. Problem: We often have multiple possible datastores (e.g., different vectorDBs,