{"slug": "ai-agents-introduction-to-llm-and-ai-terminologies", "title": "AI Agents - Introduction to LLM and AI Terminologies", "summary": "A developer explains the fundamentals of large language models, covering model weights, open-weight versus open-source models, and key parameters such as temperature, top-k, and top-p that control output randomness. The article also introduces tokenization, embeddings, and context windows as essential concepts for understanding AI agents.", "body_md": "LLM is a model, which means an equation.\n\n```\ny = mx + c\n\ny = m1x^3 + m2x^2 + m3x + m4\n```\n\nA model is actually made up of **weights**. In any model, e.g., ChatGPT model or Gemini model, they would have used a large amount of input to train the model.\n\nInput means a large amount of text/image data that is available on the internet. The input would have been fed into the **Transformer architecture** to get the output, which is the model.\n\nWeights are floating-point numbers that represent the model's learned parameters. A **10B or 100B parameter model** means how many parameters (weights) are present inside the model.\n\nWe cannot store a large-parameter model on our computer due to inadequate storage and computational power. Storage and CPU/GPU power decide what size of model can be run on a computer.\n\nTo run a model locally, we can use one of the following tools:\n\nAn **open-weight model** shares its model weights. So, we can run them, fine-tune them, and host them on a local system.\n\nHere, the training code, data, and full methodology are not shared.\n\nWhereas, in an **open-source model**, the weights, training code, data, and sometimes the dataset are shared.\n\nLLM is a **next-word predictor**.\n\nSuppose we ask:\n\n**\"Hi, how...\"**\n\nThe answer can be:\n\nThese are possibilities.\n\nHere, most of the time, the answer will be **\"How are you?\"** because if a word has more presence, it has a higher possibility of occurring.\n\nEach possibility will have a score between **0 and 1**.\n\nWe have **3 controlling parameters** to control the output generated by the LLM.\n\nUsually set from **0–1**. It controls the randomness of the model.\n\nIf the value is **0–0.3**, which is low, it means generating the most likely words, i.e., facts or commonly occurring words.\n\nIf the value is high, the model will choose less likely words.\n\nWe use this high value in **storytelling and creative writing**.\n\nControls the number of possibilities.\n\nFor example, **K = 3** means choosing only the 3 most likely possibilities.\n\nTop K will be used along with Temperature.\n\nFor example, if we set **Top K = 5** and **Temperature = 0.5**, the LLM will take possibilities from the selected Top K values based on the temperature.\n\nTop K is used to limit the number of possibilities.\n\nIt is also called a **sampling method**.\n\nThe possibility scores are added up until they reach the defined value in Top P.\n\nTop P is also used along with **Temperature**.\n\nTokens are the process of splitting words into small pieces. A small piece may not be complete.\n\nFor example, a token can be split into **\"to\"** and **\"kens\"**.\n\nThese are called **tokens**, and the process is called **tokenization**.\n\nA tokenizer, such as those used by **OpenAI and GPT**, can be used.\n\nWe can specify the size of the token that we want to split.\n\nEach token will be represented by a number, which in turn is converted into an embedding.\n\nIt is a **short-term memory**.\n\nThe context window is the maximum number of tokens that a model can see at a given time.\n\n**Short context window** and **long context window** are the types.", "url": "https://wpnews.pro/news/ai-agents-introduction-to-llm-and-ai-terminologies", "canonical_source": "https://dev.to/ramya_perumal/ai-agents-introduction-to-llm-and-ai-terminologies-1pmi", "published_at": "2026-09-01 18:38:49+00:00", "updated_at": "2026-09-01 18:54:15.532200+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence", "ai-agents", "developer-tools"], "entities": ["ChatGPT", "Gemini", "OpenAI", "GPT"], "alternates": {"html": "https://wpnews.pro/news/ai-agents-introduction-to-llm-and-ai-terminologies", "markdown": "https://wpnews.pro/news/ai-agents-introduction-to-llm-and-ai-terminologies.md", "text": "https://wpnews.pro/news/ai-agents-introduction-to-llm-and-ai-terminologies.txt", "jsonld": "https://wpnews.pro/news/ai-agents-introduction-to-llm-and-ai-terminologies.jsonld"}}