OpenAI researchers debut GPT-3 language model trained with 175B parameters, far more than GPT-2's biggest version with 1.5B parameters (Khari Johnson/VentureBeat)

Khari Johnson / VentureBeat:
OpenAI researchers debut GPT-3 language model trained with 175B parameters, far more than GPT-2's biggest version with 1.5B parameters  —  A team of more than 30 OpenAI researchers have released a paper about GPT-3, a language model capable of achieving state-of-the-art results on a range …

from Techmeme https://ift.tt/2zUSDGb