Skip to content
TheChatArchiveThe Chat Archive
Mistralmistral-medium· Mistral AIHugging Face Arena Human Preference 55K

what are parameters in an LLM?

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with gpt-4-1106-preview’s answer.
  1. User

    what are parameters in an LLM?

  2. Mistral Assistant

    In a large language model (LLM) like me, parameters refer to the internal variables that the model uses to generate responses. These parameters are learned during the training process, and they capture patterns in the data that the model was trained on.

    Specifically, in a neural network-based language model, the parameters typically include the weights and biases of the connections between the nodes in the network. During training, the model adjusts these parameters in order to minimize the difference between its predicted outputs and the actual outputs in the training data. Once the training process is complete, the parameters are fixed and can be used to generate responses to new input.

    In practice, LLMs like me can have billions or even trillions of parameters, which allows them to capture a wide range of linguistic patterns and generate high-quality responses. However, this also makes them computationally expensive to train and deploy.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.