LLM
/el el em/noun
LLM stands for large language model: an AI model trained on large amounts of text. It can process language and generate text, including answers, summaries and code.
We give the LLM our company knowledge to help it write relevant ad copy.
An LLM processes text in small units called tokens. When generating text, it uses the available context to determine which token can follow. This builds an answer step by step. Google explains how language models work in its machine learning course.
What does an LLM do with our files?
At 10KB, company knowledge, web pages and ad campaigns live in text files. We can supply these as context so the model can align its work with our services and working practices. This is an important reason for our Glossary · In briefcompany as codeCompany as code is an approach to recording company knowledge, processes and settings in text files and code. This lets you manage them with version control, checks and automation.Read more approach.
An LLM is the language model. An Glossary · In briefagentAn AI agent is software that uses a language model and tools to carry out a task. It chooses its next steps based on the goal and the results of earlier steps.Read more combines a model with tools to read files, make changes and run checks, for example. We use that combination to carry out work on our website and campaigns, then review the changes.
An LLM is not a search engine that automatically consults current sources. An application can add search or file access, but must supply that information to the model. The answer can still contain incorrect connections or invented details.
The task determines the checks required. An editor can review a draft; software can verify that processed data meets defined requirements. Test realistic situations and make clear which information the model did and did not receive.