Using Cloud TPUs for open model fine-tuning

The notebook from this video → https://goo.gle/4jAkqwp
Introduction to TPUs → https://goo.gle/4jyybvG
Setup guide for a TPU VM → https://goo.gle/4jKaoJs

Dive into the world of Cloud TPUs and learn how to use them for supervised fine-tuning of open models. Join Googlers Wietse Venema and Duncan Campbell as they focus on the Gemma model and using Hugging Face Optimum TPU. You’ll also learn how TPUs differ from CPUs and GPUs, when to choose TPUs, and how to fine-tune a model with TPUs.

More resources:
Video: Learn the difference between CPU, GPU, and TPU → https://goo.gle/4aCopVf
Video: Fine-tuning open models with TRL → https://goo.gle/4hyjcA1

Watch more Google Cloud: Building with Hugging Face → https://goo.gle/BuildWithHuggingFace
Subscribe to Google Cloud Tech → https://goo.gle/GoogleCloudTech

#GoogleCloud #HuggingFace

Speakers: Wietse Venema, Duncan Campbell
Products Mentioned: Cloud TPUs, Gemma