llama_sampler_init_softmax

Function llama_sampler_init_softmax 

Source
pub unsafe extern "C" fn llama_sampler_init_softmax() -> *mut llama_sampler
Expand description

@details Sorts candidate tokens by their logits in descending order and calculate probabilities based on logits. NOTE: Avoid using on the full vocabulary as the sorting can become slow. For example, apply top-k or top-p sampling first.