ConceptDBISAI-written · not yet proofread

Large Language Model

Also LLM · Large Language Models

A large language model (LLM) is a neural language model whose network is very large; it is used as a base model (next-token prediction) or as an instruct model that converses, steered by a system prompt.

Know these firstLanguage Model

Properties#

Context size: the number $n$ of previous tokens; very large for LLMs (400k tokens for GPT-5, 1M for Gemini Pro).
Prompt: the sequence of previous tokens. If it is longer than $n$ , only the last $n$ tokens count, or RAG selects the relevant part of a large text.
System prompt: instruct models typically have a system prompt that tells the model how to “behave”. LLM services send the concatenation of system prompt and request prompt as the tokens of a request.
Next $k$ tokens: LLM services return not just the next token but the next $k$ tokens (max_completion_tokens).
Temperature: controls how random the choice of the next token is (higher = more random).

Example#

ExampleChat-completion request (ES12)

POST https://ad-llm.cs.uni-freiburg.de/v1/chat/completions

Content-Type: application/json
Authorization: Bearer YOUR_API_KEY

{ "model": "Qwen/Qwen3-Next-80B-A3B-Instruct-FP8",
  "messages": [
    { "role": "system", "content": SYSTEM_PROMPT },
    { "role": "user", "content": PROMPT }
  ],
  "max_completion_tokens": k, "temperature": 0.7 }

In the live demo, this instruct model got the system prompt “Act like a character-based language model and predict the next characters” and was compared with the unigram, bigram and trigram models (N-Gram Model) on the same prompt.

Sources#

Slides 12, p. 10 (neural LMs, LLM), Slides 12, p. 11 (base vs. instruct), Slides 12, p. 12 (terminology), Slides 12, p. 13 (API), Slides 12, p. 14 (context sizes)
DBIS_VL12 > Neural and large language models, DBIS_VL12 > Base models vs. instruct models, DBIS_VL12 > Terminology, DBIS_VL12 > Using an LLM via API (ES12)

Connections

Prerequisites know first1

Language Model

Mentioned in inline links4

Language Model
A neural language model with a very large network is a large language model (LLM). All of these, including LLMs, are “the same kind of model”, only trained and computed differently. §
Base vs. instruct models, system prompt, temperature and the API: see Large Language Model. §
Retrieval-Augmented Generation
Retrieval-augmented generation (RAG) lets an LLM answer a query over a text that exceeds its context size: based on the query, extract a small subset of the text and pass only that to the LLM, together with a suitable prompt. §
VL12 – N-gram Models, LLMs, RAG
Large Language Model — neural language model with a very large network §
Word Embeddings and Language Models
Large Language Model — very large neural language model; base vs. instruct, system prompt, API §
LLM: a very large neural LM. A base model only continues text; an instruct model is trained to converse and is steered by a system prompt. Services return $k$ tokens per request (max_completion_tokens, temperature). §

Search in these mentions →