Google has recently introduced a trio of new “open” generative AI models, namely Gemma 2 2B, ShieldGemma, and Gemma Scope. Thee models will focus on safety. In addition, they are part of Google’s Gemma 2 family of generative models.
Let’s delve into the details of these models and their intended applications.
Gemma 2 2B by Google
Gemma 2 2B is a lightweight model for generating and analyzing text. It is capable of running on various hardware. Additionally, it includes laptops and edge devices. There are specific research and commercial applications licensed for this model. In addition, users can obtain them from sources such as Google’s Vertex AI model library, Kaggle, and Google’s AI Studio toolkit.
ShieldGemma
ShieldGemma is a collection of “safety classifiers.” In addition, it aimed at detecting toxicity, including hate speech, harassment, and sexually explicit content. Based on Gemma 2, it is an extension of Gemma. Additionally, users can utilize it to filter prompts to a generative model and the content generated by the model.
Gemma Scope by Google
Gemma Scope allows developers to gain a deeper understanding of specific points within a Gemma 2 model. It comprises specialized neural networks. Additionally, it helps unpack the complex information processed by Gemma 2, making it easier to analyze and comprehend. By studying these expanded views, researchers can gain valuable insights into how Gemma 2 identifies patterns, processes information, and makes predictions.
Closing Note
These new models are part of Google’s effort to foster goodwill within the developer community, similar to Meta’s attempt with Llama. The release of these models aligns with the U.S. Commerce Department’s endorsement of open AI models. As a result, it emphasizes their potential to broaden availability to smaller companies, researchers, nonprofits, and individual developers. However, it highlights the need for capabilities to monitor such models for potential risks.
Also Read: https://thecitizenscoop.com/floorr-unveil-new-styling-tool-for-stylists/
The introduction of these models reflects Google’s commitment to advancing the field of generative AI. However, it prioritizes safety and transparency in their applications.
