The hybrid Jamba architecture enables Jamba-1.5 models to reach excellent throughput and latency, especially at long contexts. With the same hardware, Jamba-1.5 models are the fastest across the board (in the image: 2xA100 80GB GPUs for Mini, 8xA100 80GB GPUs for Large). 2/7
GENERATIVE AI
-
Jamba-1.5 Whitepaper Released: Hybrid SSM-Transformer Models
By
–
Jamba-1.5 whitepaper is out!
The whitepaper details the architecture, training schemes, novelties and in-depth evaluations of our new long context hybrid SSM-Transformer models – Jamba-1.5-Large and Jamba-1.5-Mini. Arxiv: https://
arxiv.org/abs/2408.12570 Here are some highlights and -
Phi-3.5: Microsoft Introduces Its New Generation of SLMs
By
–
[#Article] Phi-3.5: @Microsoft presents the latest generation of its SLMs, optimized for specific tasks https://actuia.com/actualite/phi-3-5-microsoft-presente-la-derniere-generation-de-ses-slm-optimisee-pour-des-taches-specifiques/
… #AI #ArtificialIntelligence -
Claude Family Models Excel with Strong Alignment Safety Measures
By
–
The Claude family from @AnthropicAI is really good at it: Claude 3.5 Sonnet, 3 Opus & Haiku all succeed. It feels like a heavy alignment stage with refusals is part of the explanation.
-
GPT-4 Outperforms 4o Versions in Question Answering
By
–
The old GPT-4 also correctly answers the question, while the 4o versions fail.
-
AI Model Hallucinations: Size Doesn’t Guarantee Factual Accuracy
By
–
The "Indigo Sock Game" doesn't exist but most models will hallucinate it for you. Factuality hallucinations are fascinating because they often behave in unexpected ways. You'd think bigger models would perform better, but that's not necessarily true.
-
Open-Source LLMs: Unsung Pioneer in AI Evolution
By
–
Doesn't get enough credit but IMO paved the way for open-source LLMs!
-
Diffusion Models as Autoregressive Models: Understanding Frequencies
By
–
The blogpost: https://
sander.ai/2023/07/20/per
spectives.html#:~:text=Diffusionmodelsareautoregressivemodels,-Fordiffusionmodels&text=Tounderstandthisithelps,detailscorrespondtohighfrequencies
…. -
Google Expands AI Overviews to Six New Countries with Local Language Support
By
–
Google is taking its AI-powered Search to the next level by expanding AI Overviews to six new countries. India, the United Kingdom, Japan, Indonesia, Mexico, and Brazil will have local language support along with AI Search!
Starting with the U.S., Google’s innovative AI Overviews