Published
Sparse Autoencoders (SAEs) are an interpretability tool that allow you to inspect what concepts your LLM is thinking about. Learn more our open-source suite of SAEs on all layers of Gemma 2 2B and 9B, which enables everyone to analyze and steer Gemma's internals.
Watch more videos from Gemma Developer Day Tokyo → https://goo.gle/gemma-dev-tokyo
Subscribe to Google for Developers → https://goo.gle/developers
#Gemma #GemmaDeveloperDay
Watch more videos from Gemma Developer Day Tokyo → https://goo.gle/gemma-dev-tokyo
Subscribe to Google for Developers → https://goo.gle/developers
#Gemma #GemmaDeveloperDay
- Category
- Project
- Tags
- Google, developers
Sign in or sign up to post comments.
Be the first to comment
Up Next
-
09:37
Demo: Using Gemma with the Hugging Face ecosystem
-
Demo: Gemma 2 2B on a Jetson Orin Nano
-
Demo: Teach Gemma to work in YOUR language
-
Demo: Gemma on-device with MediaPipe
-
11:34
Demo: Taking Gemma from prototype to production faster with Vertex AI
-
11:37
Demo: Rapid prototyping with Gemma and Llama.cpp
-
What's new in Gemma 2
-
Demo: Multi-modal Gemma: PaliGemma with Keras
-
08:12
Demo: JAX, Flax and Gemma
-
Demo: Gemma data, training, and the path to improvement
-
Demo: Post-training research with Gemma
-
Demo: Gemma 2 architecture: JAX, Flax, and more
-
26:21
Hands-on introduction to Gemma
-
09:00
Getting started with Gemma models
-
10:07
Demo: Building a Gemma chatbot with Keras
-
12:21
Demo: Optimizing Gemma inference on NVIDIA GPUs with TensorRT-LLM
-
What Gemma model can help me code?
-
08:34
Demo: Deploying Gemma at dataflow scale
-
Demo: Unleashing Gemma in production with Hugging Face Text Generation Inference (TGI)
-
11:51
Gemma: The responsible way to build
-
07:57
Demo: Gemma on-device with MediaPipe and TensorFlow Lite
-
00:15
Run AI locally, with Gemma!
Loading...