Below are some extracts from the paper about GPT-3's performance: (link to paper) https://
arxiv.org/abs/2212.09196
v1
…
RESEARCH
-
Extracts from GPT-3 Performance Paper
By
–
-
GPT-3 Outperforms on IQ Test in UCLA Study
By
–
How does GPT-3 perform on an actual IQ test? UCLA published findings on how various Language Models performed on analytical reasoning questions against an average college applicant. The AI was the clear winner.
-
Data Growth and Machine Learning: The Human Question Challenge
By
–
The rise of machines! Data predicted to grow 5x by 2025 >>> yet, humans need to ask and action the right questions >>> chart by @Applied4Tech via @MikeQuindazzi >>> #DataScience #MachineLearning #Sensors #DataAnalytics #AI #IoT
-
Progressive Training Strategy for Photorealistic Image Generation Models
By
–
Exactly! Train on photorealistic first then use that base to train styles and then use that base for img2img or even Dreambooth. Endless possibilities to try
-
Gary Marcus and Vincent Boucher Host Major AI Debate
By
–
Gary Marcus and Vincent Boucher hosted another of their epic AI debates with a cast of fascinating speakers. // #agidebate @Montreal_AI @GaryMarcus // #AI #machinelearning #deeplearning
-
OpenAI Technology: Anticipating Problems and Iterative Improvement
By
–
in particular, there are going to be significant problems with the use of openai tech over time; we will do our best but will not successfully anticipate every issue. we will run a very tight feedback loop for improvement, and try to make our mistakes while the stakes are low.
-
Technological Progress Drives Sustainable Economic Growth
By
–
technological progress is the only thing that delivers real, sustainable economic growth
-
Charles Babbage’s Difference Engine: Computing Pioneer
By
–
Happy birthday to Charles Babbage, who pioneered computing with his "Difference Engine." http://
bit.ly/1MvReDN -
Why Scaling Large Language Models Matters in AI
By
–
A estas tecnologías se le llaman Enormes Modelos del Lenguaje (LLMs) y si queréis saber por qué es tan interesante hacer a estas IAs cada vez más grandes, en este vídeo se explica muy bien 🙂
-
Chinchilla shows scaling laws matter more than model size
By
–
Y es que este año DeepMind con Chinchilla demostró que no hacía falta escalar tanto a estos Enormes Modelos del Lenguaje. Que GPT-3 todavía tenía margen de mejora para ser entrenado más y con más datos sin necesidad de hacerlos MÁS GRANDES. Así que el tamaño no es tan importante.