Skip to main content

​
FunctionGemmaConfig

class FunctionGemmaConfig(BaseConfig):
Defines the parameters for generating completions using FunctionGemma via Ollama’s native API. FunctionGemma uses a custom chat template format for function calling that differs from OpenAI’s format. This config is used with Ollama’s /api/generate endpoint. Reference: https://github.com/ollama/ollama/blob/main/docs/api.md Parameters:
  • temperature (float, optional): Sampling temperature to use, between :obj:0 and :obj:2. Higher values make the output more random, while lower values make it more focused and deterministic. (default: :obj:None)
  • top_p (float, optional): An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. (default: :obj:0.95)
  • top_k (int, optional): Limits the next token selection to the K most probable tokens. (default: :obj:64)
  • num_predict (int, optional): Maximum number of tokens to generate. (default: :obj:None)
  • stop (list, optional): Sequences where the model will stop generating further tokens. (default: :obj:None)
  • seed (int, optional): Random seed for reproducibility. (default: :obj:None)