What is a base model?
A pre-trained model that can serve as the starting point for fine-tuning to address specific tasks or applications.
base model explained in plain English
A pre-trained model that can serve as the starting point for fine-tuning to address specific tasks or applications. See also pre-trained model and foundation model.
Example
Practitioners refer to base model when building, training, or evaluating machine learning systems. It appears in research papers, product documentation, and technical discussions about AI capabilities and limitations.
People also read
- clipping
A technique for handling outliers by doing either or both of the following: - Reducing feature values that are greater than a maximum threshold down to that maximum threshold.
- foundation model
A very large pre-trained model trained on an enormous and diverse training set.
- Gemini
The ecosystem comprising Google's most advanced AI.
- GPT
A family of large language models developed by OpenAI that generate human-like text.
- large language model
At a minimum, a language model having a very high number of parameters.
- Low-Rank Adaptability
A parameter-efficient technique for fine tuning that "freezes" the model's pre-trained weights (such that they can no longer be modified) and then inserts a small set of trainable weights into the model.
- pre-trained model
Although this term could refer to any trained model or trained embedding vector, pre-trained model now typically refers to a trained large language model or other form of trained generative AI model.
- retrieval-augmented generation
A technique for improving the quality of large language model (LLM) output by grounding it with sources of knowledge retrieved after the model was trained.
- tokenizer
A system or algorithm that translates a sequence of input data into tokens.
- auto-regressive model
A model that infers a prediction based on its own previous predictions.