Ah yeah! I bet you can get massive benefits just by upgrading the stack to Qwen 2.5
LLMS
-
Llama PDF Pre-Processing Logic Guide Released
By
–
You can find it here: https://
github.com/meta-llama/lla
ma-recipes/blob/main/recipes/quickstart/NotebookLlama/Step-1PDF-Pre-Processing-Logic.ipynb
… -
Meta Releases Llama Recipes GitHub Repository for Quick Start
By
–
Check it out here: https://
github.com/meta-llama/lla
ma-recipes/tree/main/recipes/quickstart/NotebookLlama
… -
Meta Releases NotebookLlama: Open Recipe Using Llama Models
By
–
Wow! Meta dropped an open NotebookLM recipe: NotebookLlama 🔥
— Vaibhav (VB) Srivastav (@reach_vb) 27 octobre 2024
It uses L3.2 1B/ 3B for pre-processing the PDF, L3.1 70B for Transcript creation, L3.1 8B for re-writes and Parler TTS for Text to Speech ⚡
Step 1: Pre-process PDF: Use Llama-3.2-1B-Instruct to pre-process the PDF… pic.twitter.com/L7hb5GsMtlWow! Meta dropped an open NotebookLM recipe: NotebookLlama It uses L3.2 1B/ 3B for pre-processing the PDF, L3.1 70B for Transcript creation, L3.1 8B for re-writes and Parler TTS for Text to Speech Step 1: Pre-process PDF: Use Llama-3.2-1B-Instruct to pre-process the PDF
-
The Rarity of Sane LLM Whisperers for Useful Commentary
By
–
The problem is that to get any useful commentary we'd need to find an LLM Whisperer who is not insane. I'm given to understand that they exist, but they're rare.
-
Strange Bimodal Results: The Role of Exact Prompts
By
–
Okay, then what oneshot prompt? We've really got a strange bimodal thing going on where some people report that they can't get any good results, and it probably has something to do with exact prompts!
-
William Guy Lecturers 2024-25 AI and Machine Learning Series
By
–
Current William Guy Lecturers 2024-25 https://
bit.ly/4gRscRt
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
OpenAI’s Orion Model Launch Announcement Coming Soon
By
–
It's Official: OpenAI’s Orion is Just Around the Corner—And It’s Set to … https://
youtu.be/mvccfCwIYY4?fe
ature=shared
… via @YouTube -
Cerebras Sets New Speed Record for Llama 3.1-70B Serving
By
–
Congrats @andrewdfeldman and @CerebrasSystems for a huge leap forward and setting a new speed record for serving Llama 3.1-70B. 2100 tokens/sec is blazingly fast for a 70B model. This is great for agentic AI! https://t.co/LmIAXKQb3W
— Andrew Ng (@AndrewYNg) 26 octobre 2024Congrats @andrewdfeldman and @cerebras for a huge leap forward and setting a new speed record for serving Llama 3.1-70B. 2100 tokens/sec is blazingly fast for a 70B model. This is great for agentic AI!
-
What Matters for Model Merging at Scale?
By
–
What Matters for Model Merging at Scale? Yadav et al.: https://
arxiv.org/abs/2410.03617 #ArtificialIntelligence #DeepLearning #MachineLearning