cyberlabe 🦉 (@cyberlabe) profile picture
Lille, France
Joined April 2008
Photo by cyberlabe 🦉 (@cyberlabe)Photo by cyberlabe 🦉 (@cyberlabe)Photo by cyberlabe 🦉 (@cyberlabe)Photo by cyberlabe 🦉 (@cyberlabe)Photo by cyberlabe 🦉 (@cyberlabe)Photo by cyberlabe 🦉 (@cyberlabe)
New Research: Cascaded Semantic Fractionation for identifying a domain in social media: Searching social media to find relevant semantic domains often results in large text files, many of which are irrelevant due to cross-domain content… frontiersin.org/articles/10.… #ResearchMetrics
1
113
Midjourney, DALL•E 3 and GPT-4 have opened a world of endless possibilities. I just coded "Angry Pumpkins 🎃" (any resemblance is purely coincidental 😂) using GPT-4 for all the coding and Midjourney / DALLE for the graphics. Here are the prompts and the process I followed:
520
1,793
702
11,368
5,385,167
cyberlabe 🦉 retweeted
It’s weird to sit in a restaurant or something and look around and realize nobody knows what’s about to happen with AI. It’s like you know aliens exist and nobody else knows.
106
134
32
1,147
275,375
Un passionnant nouveau chapitre de l'IA : découvrez les agents mathématiques de l'IA, qui analysent, écrivent, résolvent et découvrent des équations, créant ainsi des écosystèmes mathématiques fascinants ! 🌐 #IA #Mathématiques slideshare.net/lablogga/ai-m…
46
cyberlabe 🦉 retweeted
Introducing BTLM-3B-8K: an open, state-of-the art 3B parameter model with 7B level performance. When quantized, it fits in as little as 3GB of memory 🤯. It runs on iPhone, Google Pixel, even Raspberry Pi. BTLM goes live on Bittensor later this week! 🧵👇 buff.ly/3Q5dtY5
19
161
36
518
147,533
4-bit QLoRA is here to equalize the playing field for LLM exploration. You can now fine-tune a state-of-the-art 65B chatbot on one GPU in 24h. Paper: arxiv.org/abs/2305.14314 Code and Demo: github.com/artidoro/qlora
QLoRA: 4-bit finetuning of LLMs is here! With it comes Guanaco, a chatbot on a single GPU, achieving 99% ChatGPT performance on the Vicuna benchmark: Paper: arxiv.org/abs/2305.14314 Code+Demo: github.com/artidoro/qlora Samples: colab.research.google.com/dr… Colab: colab.research.google.com/dr…
6
46
4
220
81,437
cyberlabe 🦉 retweeted
3/n Here are functions available for browsing mode
1
4
65
44,629
I built a ChatGPT app that lets you chat with any codebase! 99% of projects just copy/paste Langchain tutorials. This goes well beyond that. Here's how I built it:
45
194
16
1,606
484,223
BabyAgi following a detailed description of a small Python program and correctly creating it by managing multiple files with different classes. In this version, it can write new code, edit existing code and run commands (e.g., to install dependencies) @yoheinakajima @babyAGI_
10
36
10
225
50,485
cyberlabe 🦉 retweeted
«les adolescents et jeunes adultes rencontrés se préoccupent grandement de la qualité de l’information à laquelle ils sont confrontés, que ce soit en termes de contenu ou en termes de fiabilité» Anne Cordier cfeditions.com/grandir-infor… en précommande Please RT
1
15
3
15
4,608
cyberlabe 🦉 retweeted
The Personal Knowledge Graphs book is coming. The Kindle pre-order will be launched soon. Stay tuned. Subscribe at PersonalKnowledgeGraphs.com, so you get an email when the pre-order opens.
2
14
4
65
7,738
cyberlabe 🦉 retweeted
We are launching the first 100% 🇪🇺 European public DNS resolver! Free, sovereign and operated by an independent non-profit organization based in France. Thanks to @NextDNS, @bambenek, @BforeAi, @abuse_ch, @_ANEXIA, @gandibar who helped us launch! [1/3] dns0.eu
26
168
25
277
77,241
Les nouvelles fonctionnalités Google qui pourraient arriver en France prochainement bases-netsources.com/breves-…
27
J'ai testé l'utilisation de ChatGPT pour l'analyse de contenu... une démarche intéressante pour accompagner l'enseignement à l'esprit critique.
1
84
Pourquoi les nouveaux moteurs de recherche qui utilisent de grands modèles de langage donnent des réponses imparfaites garymarcus.substack.com/p/is…
1
83