Input hooks are special pieces of code that allow a program to watch for and react to specific user inputs, such as keyboard presses or mouse movements. They act like listeners, waiting for certain actions so that the software can respond immediately. This mechanism is often used to customise or extend how a program handles…
Token usage refers to the number of pieces of text, called tokens, that are processed by language models and other AI systems. Tokens can be as short as one character or as long as one word, depending on the language and context. Tracking token usage helps manage costs, performance, and ensures that the input or…
Prompt efficiency refers to how effectively and concisely a prompt communicates instructions to an AI system to get accurate and relevant results. It involves using clear language, avoiding unnecessary details, and structuring requests so the AI can understand and respond correctly. Efficient prompts save time and resources by reducing the need for repeated clarifications or…
Prompt polishing is the process of improving and refining instructions given to an AI model to achieve clearer, more accurate, or more relevant responses. It involves editing the wording, structure, and detail of prompts so the AI better understands what is being asked. Effective prompt polishing helps avoid misunderstandings and increases the quality of the…
Tone control refers to the ability to adjust the balance of different frequencies in an audio signal, such as bass, midrange, and treble. It allows users to make the sound warmer, brighter, or more balanced according to their preferences or the acoustics of a room. Tone controls are commonly found on audio equipment like amplifiers,…
Response filters are tools or processes that modify or manage the information sent back by a system after a request is made. They can check, change, or enhance responses before they reach the user or another system. This helps ensure that the output is correct, safe, and meets certain standards or requirements.
Prompt sanitisation is the process of checking and cleaning user input before it is sent to an AI system or language model. This step helps to remove harmful, inappropriate or malicious content, such as offensive language, private information or code that could be used for attacks. It ensures that prompts are safe, appropriate and do…
Prompt regression refers to a gradual decline in the effectiveness or accuracy of responses generated by an AI language model when using a specific prompt. This can happen when updates to the model or system unintentionally cause it to interpret prompts differently or produce less useful answers. Prompt regression is a concern for developers who…
Prompt templates are pre-written structures or formats used to guide artificial intelligence systems, especially language models, in generating consistent and relevant responses. They help users by providing a starting point or framework for questions, instructions, or tasks, reducing the need to write prompts from scratch each time. This approach improves efficiency and ensures that AI…
Prompt routing is the process of directing user prompts or questions to the most suitable AI model or system based on their content or intent. This helps ensure that the response is accurate and relevant by leveraging the strengths of different models or tools. It is often used in systems that handle a wide variety…