Few-shot learning with GPT-J and GPT-Neo

 Hello!

Since I added the GPT-J and GPT-Neo endpoints to NLPCloud.io, I've had many questions about how to make the most of these 2 great NLP models.
So I just wrote an article about few-shot learning with GPT-J and GPT-Neo: a simple technique to dramatically improve accuracy:
https://nlpcloud.io/effectively-using-gpt-j-gpt-neo-gpt-3-alternatives-few-shot-learning.html

Few-shot learning is about helping a machine learning model make predictions thanks to only a couple of examples. No need to train a new model here: models like GPT-J and GPT-Neo are so big that they can easily adapt to many contexts without being re-trained.

Thanks to this technique, I'm showing how you can easily perform things like sentiment analysis, code generation, tutorial generation, machine translation, spell correction, question answering, tweet creation…

I personally find it amazing what can be done with these NLP models. It seems that only our imagination is the limit!

Hope you'll find it useful.