Advanced Search
Search Results
48 total results found
Poolnoodle
As a floatation device A pool noodle is a long, buoyant, and lightweight foam tube, typically used as a flotation device in swimming pools. It's often used for fun, relaxation, and safety, especially for children or non-swimmers. The noodle floats on water, p...
Summary persona
This persona is normally not visible for the public. But, you use it every day. This persona is used to generate a short descriptive title for a chat session, based on the first prompt the user enters. The persona is used by ScaiWave internally to give title...
Top P
The Top P value is a parameter used in AI models, particularly in natural language processing (NLP) models like GPT-3 or our own Poolnoodles, to control the diversity of generated text. It is a probability threshold that determines the likelihood of a word or ...
Temperature
In the context of AI models, particularly in natural language processing (NLP) models, the temperature setting is a parameter that controls the randomness or creativity of the generated text. It is a numerical value between 0 and 1, with a lower temperature re...
Under Construction
Persona University is still being developed. As soon as it is done, this chapter will be updated.
Context length
The maximum number of tokens or characters that a language model can process and consider in a single input. It determines how much "context" the model can take into account when generating responses or making predictions. Exceeding the context length may requ...
RAG (Retrieval-Augmented Generation)
A method in natural language processing that combines traditional large language models (LLMs) with external information retrieval systems. In RAG, the model generates responses by first retrieving relevant information from a database or document collection an...
Token
In the context of natural language processing (NLP), a token is the smallest unit of text that a language model processes. Tokens can be words, subwords (smaller units within words), or even characters, depending on the model's tokenization method. The number ...