Abliterating LLMs is the most interesting trend I've seen in months A simple weight modification can jailbreak models without any retraining. Here's how it works: Identification – Run model on harmful & harmless prompts
– Capture activations at the last token position
–
LLMS
-
Weight Modification Jailbreak Technique for Large Language Models
By
–
-
AI Struggles with Factual Accuracy in Search Applications
By
–
I guess my primary use case for Google is to search up facts, as in: "I dunno, lemme Google it". Turns out this primary use case, getting facts right, is not a strength for today's AI.
-
Questioning Whether LLMs Are Optimal for World Modeling
By
–
Makes you wonder if LLMs are the best way to model the world at all…
-
Using AI Prompts for Eisenhower Matrix Task Prioritization
By
–
3. Prioritize tasks effectively: Prompt: "List my tasks for [specific day or project] and help me prioritize them based on the Eisenhower Matrix of urgent vs. important."
-
Local RAG App with PDF Support Using Llama-3
By
–
You can build a local RAG app that works with PDF and runs on Llama-3 deployed locally.
-
Free LLM Version Coming with Llama-3 and Ollama
By
–
That has become very minimal especially after gpt-4o and I never said it's free to use. The free version is coming soon that will run on locally deployed llama-3 with @ollama
-
LLM Tips and Tutorials for Sharing
By
–
If you find this useful, RT to share it with your friends. Don't forget to follow me @Saboo_Shubham_ for more such LLMs tips and tutorials.
-
Awesome LLM Apps with RAG GitHub Repository
By
–
Find all the awesome LLM Apps demo with RAG in the following Github Repo. P.S: Don't forget to star the repo to show your support
