Gemma
Gemma is a family of open-weight generative AI models developed by Google.
Deep Learning with R, Third Edition
Gemma is a family of pretrained generative language models released in specialized instruction-tuned variants.
- How it is used
- Deploys a pretrained generative language model for text generation and image-and-text prompting.
- How deep it goes
- Explains loading, tokenization, generation, next-token prediction, and image soft tokens, but stops short of detailing fine-tuning.
- Examples
- Demonstrates advice, brownie recipes, an impossible-president question, image descriptions, and visual question answering.
- What you can do after
- Operate/configure Gemma through KerasHub, generate prompted continuations, and submit image-and-text inputs.
Read how Deep Learning with R, Third Edition explains Gemma →
Deep Learning with Python, Third Edition
A pretrained large language model family used here for text generation and later instruction fine-tuning.
- How it is used
- Uses a pretrained generative model to demonstrate text completion before addressing more useful instruction-oriented behavior.
- How deep it goes
- Explains model scale, loading, tokenization, generation, sampling, and pretrained limitations, but stops short of training internals.
- Examples
- Demonstrates advice completion, brownie instructions, and a fabricated answer about a nonexistent president.
- What you can do after
- Operate/configure Gemma through KerasHub, authenticate for its weights, generate continuations, and recognize unsupported factual output.
Read how Deep Learning with Python, Third Edition explains Gemma →
Build Applications with Local AI Models on a Mac
Gemma is a family of open language models developed by Google.
- How it is used
- Supports practical comparison and selection of models for applications such as a chatbot.
- How deep it goes
- Compares release sizes and capabilities and gives an efficiency example, without detailing model internals.
- Examples
- Compares Gemma release sizes and cites Gemma 2 9B as an example of efficiency.
- What you can do after
- Evaluate and select models by size, speed, quality, and task suitability.
Read how Build Applications with Local AI Models on a Mac explains Gemma →
Related concepts
- language modelkind of · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
- KerasHubused by · KerasHub is a Keras library providing reusable components, pretrained models, and utilities for machine-learning and generative-modeling tasks.
- Qwencontrasts with · Qwen is a family of pretrained generative language models developed by Alibaba Cloud, with variants for different tasks and modalities.
- Ollamauses · Ollama is software for downloading, running, and serving language models locally.
- open-weight modelkind of · An open-weight model is a machine-learning model whose trained parameter values are made available for others to download, run, or modify, subject to applicable terms.
- model familykind of · A model family is a group of related models that share characteristics such as architecture, design, or training lineage.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.