Member-only story
5 things you should know about OpenAI o1
Last week openai launched o1 series. They dominated every headlines of newsletters I received. In this article I will point out some significant things that I noticed and thought to be true.
The o1 models are specially trained to do the task of reasoning and thinking. Something that OpenAI has beed interested since long. Here are some intersting takeways I wanted to share with you.
1. Training
OpenAI utilized Reinforced learning (RL) specifically optimized for chain-of-thought(cot) scenarios. Chain-of-Thought is a recent advancement in prompting methods that encourage Large Language Models (LLMs) to explain their reasoning. This method contrasts with standard prompting by not only seeking an answer but also requiring the model to explain its steps to arrive at that answer. This is the first time OpenAI has used this method.
2. Efficient at Multilingual task
