llm

Definition

Large Language Model: a model pre-trained on large text corpora to predict the next token, then post-trained to follow instructions and behave acceptably. "Large" has no fixed threshold — the label tracks the recipe and the resulting generality rather than a parameter count. The same architecture applied to code, protein sequences, or actions is often still called an LLM by convention.

Explainers using this term

• Paper Explained: Disaggregated Quantization — Quantize LLM Prefill and Decode DifferentlyDisaggregated Quantization: Specializing LLM Prefill and Decode
• Paper Explained: Just-in-Time Memory — Curate Agent Memory When You Read It, Not When You Write ItJust-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents
• Paper Explained: Steering Geometry — Do LLM Steering Vectors Trace the Map of Human Values?Steering Geometry: Validating Human Value Geometry in LLM Steering Space