We're thrilled to announce a new course on Reinforcement Learning from Human Feedback (RLHF) built in collaboration with Google Cloud.
Large language models (LLMs) are trained on human-generated text, but additional methods are needed to align an LLM with human values and preferences. Reinforcement Learning from Human Feedback (RLHF) is a key method for aligning LLMs to make them more helpful, honest, and safe.
In this course, you will gain a conceptual understanding of the RLHF training process, and then practice applying RLHF to tune an LLM. You will:
Explore the two datasets (“preference” and “prompt”) that are used in RLHF training.
Use the open source Google Cloud Pipeline Components Library to fine-tune the Llama 2 model with RLHF.
Assess the tuned LLM against the original base model by comparing loss curves and using the “Side-by-Side (SxS)” method.
Join instructor Nikita Namjoshi, Developer Advocate for Generative AI at Google Cloud, to learn this exciting technique you can use in building your own applications.
%20(1200%20%C3%97%20200%20px)%20(1).png?upscale=true&width=602&upscale=true&name=Email%20Header%20Options%20(1200%20%C3%97%20300%20px)%20(1200%20%C3%97%20200%20px)%20(1).png)
.png?upscale=true&width=1120&upscale=true&name=DeepLearning1_GoogleCloudPlatfomr_Banner2_2070x1080%20(1).png)
.gif?upscale=true&width=1120&upscale=true&name=Untitled%20design%20(4).gif)
