LLM (large language model)
Türkçe: LLM (büyük dil modeli)
A neural network trained on very large amounts of text that generates text by probabilistically predicting the next piece (token) of a sequence.
// EXPLANATION
LLMs consist of billions of parameters and are built on the Transformer architecture. Rather than "understanding" meaning like a human, they pick a likely continuation based on patterns in the training data; so they can also produce fluent but wrong ("hallucinated") answers.
// ALSO KNOWN AS
LLMlarge language modellarge language modelslanguage model
// RELATED TERMS
TokenThe smallest unit into which a language model splits text; it can be a word, part of a word or a punctuation mark.
Transformer architectureThe neural-network architecture introduced in 2017 that computes relationships between all words in a text in parallel using an attention mechanism.
HallucinationWhen a language model presents information that does not exist as though it were true, in a confident tone.
InferenceThe stage in which a trained model produces output for new input; the phase that serves users after training.
QuantizationA technique that lowers memory and speed cost by storing model weights with fewer bits (for example 8 or 4 instead of 16).
