Mistral’s New Model Crushes Benchmarks in 4+ Languages: Mixtral 8x22B by Mistral AI Crushes Benchmarks in 4+ Languages The post… https://
analyticsvidhya.com/blog/2024/04/m
ixtral-8x22b-by-mistral-ai/?utm_source=dlvr.it&utm_medium=twitter
… #DataAnalytics #DataScience #DataDriven #ArtificialIntelligence #Infrastructure #MachineLearning #BusinessIntelligence #CTO
MACHINE LEARNING
-
Mixtral 8x22B Crushes Benchmarks Across Four Languages
By
–
-
Merge Results Expected at 90 Percent Performance
By
–
It would be interesting to see the results. I expect this merge to be like 90% as good as the full merge
-
NeurIPS 2021 Research Paper Presentation
By
–
This was presented at NeurIPS2021:https://t.co/N6yD9VTFfw https://t.co/TvgpsR2whn
— hardmaru (@hardmaru) 1 juin 2024This was presented at NeurIPS2021: https://
openreview.net/forum?id=lM297
1LAwV
… -
Incomprehensibility of SGD-Generated Programs
By
–
You can't even understand all the program that stochastic gradient descent creates.
-
AlphaGo and GPT-5: The Role of Search in AI Systems
By
–
AlphaGo = Convnets + Search
GPT-5 = GPT-4 + Search -
Better reasoning reduces AI data requirements exponentially
By
–
Better reasoning exponentially reduces the amount of data you need.
-
Quality reasoning matters more than data quantity
By
–
We don't need more data, we need better reasoning.
-
Weak Supervision Revolutionizing AI Training Data Generation
By
–
Want to see the future of weak supervision in AI? Join esteemed professors and researchers from Brown University, University of Wisconsin-Madison, and Stanford University shed light on how weak supervision is revolutionizing the way we generate AI training sets. This panel
-
Understanding Giant Tensors and Writing Goodhart-Proof Objectives
By
–
It doesn't need to be representing new physics, for us to have trouble understanding what the giant inscrutable tensors mean — or writing airtight Goodhart-proof objectives that operate over them, a much higher requirement of proficiency than current interpretability efforts.
-
Positional Encoding: Essential Beyond Attention in Transformers
By
–
Attention is not all you need. Without positional encoding, a transformer would treat a context as a bag of words.