Great resource on quantization: a complete, visual guide with beautiful graphs! 4/Int4, would recommend. Thanks a lot @MaartenGr
! https://
newsletter.maartengrootendorst.com/p/a-visual-gui
de-to-quantization
…
@aymericroucher
-
A Comprehensive Visual Guide to AI Model Quantization
By
–
-
Estimating the Cost of Frontier LLM Research and Model Training
By
–
How much does frontier LLM research cost? If you want to build models from scratch, it's about 13M USD This was calculated on a recent Google Deepmind paper by an engineer : read their full article here https://
152334h.github.io/blog/scaling-e
xponents/
… -
Building applications with Transformer agents and Llama 3.1
By
–
I made it using Transformers agents. You can find all the code under "files" in the Space, in file "
http://
app.py"
It uses llama3.1-70B, which would be too heavy to run on a consumer machine, but you can use our API as I did in the space! -
Using Llama 3.1 as an AI Data Analyst Agent
By
–
Llama 3.1 is impressive as a Data analyst Agent!
— m_ric (@AymericRoucher) 25 juillet 2024
Drop it your data as a csv, let it do the work, and you will get cool insights and plots in a matter of seconds!
Check out the demo here: https://t.co/LBaNmkOclA pic.twitter.com/svkuRRQQlyLlama 3.1 is impressive as a Data analyst Agent! Drop it your data as a csv, let it do the work, and you will get cool insights and plots in a matter of seconds! Check out the demo here: https://
huggingface.co/spaces/m-ric/a
gent-data-analyst
… -
Llama 3.1 Data Analyst Agent Demo
By
–
Llama 3.1 is impressive as a Data analyst Agent!
— m_ric (@AymericRoucher) 25 juillet 2024
Check out the demo here: https://t.co/LBaNmkNEw2 pic.twitter.com/eABJ0mUdcRLlama 3.1 is impressive as a Data analyst Agent! Check out the demo here: https://
huggingface.co/spaces/m-ric/a
gent-data-analyst
… -
Hugging Face Transformers v4.43 Update Enhances AI Agent Planning
By
–
Transformers v4.43 has been released, with a few sizeable improvements to Agents! Planning step: the agent can draw and update a plan, which considerably improves task solving trajectories. Release notes here: https://
github.com/huggingface/tr
ansformers/releases
… TY @LysandreJik for the reviews! -
User reaction to Meta’s new open-source model with increased context length
By
–
Looking forward to see what we can do with this, especially the increaseed context length! Thanks a lot to @Meta for open-sourcing this
-
Meta Releases Llama-3.1 with 405B Model and Expanded Context
By
–
Mark Zuckerberg just announced the GPT-4 killer: Llama-3.1 Two breakthroughs:
New 405B, possibly the strongest LLM ever, slightly above GPT-4o in many domains
Improved 8B & 70B models, with a much larger context size of 128k vs 8k ⇒ game-changer for RAG and Agents. -
DoLa: New AI Decoding Method Reduces Hallucinations
By
–
DoLa was just merged in transformers! This new decoding method works by contrasting token logits between the final layer and earlier layers, with the premise that high level knowledge builds in top layers rather than the first ones. It significantly reduces hallucinations!
-
Using Transformer ReactCodeAgent with Llama-3-70B for Data Analysis
By
–
I'm currently exploring Transformer's ReactCodeAgent as a data analyst on Kaggle's Titanic dataset, using Llama-3-70B-Instruct as the engine. The results are insane! The code and plot below were all generated by the agent.