Method __call__

llava/language.py:198–205

(
        self,
        inputs: mx.array,
        cache=None,
        inputs_embeds=None,
    )

        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)

    def __call__(
        self,
        inputs: mx.array,
        cache=None,
        inputs_embeds=None,
    ):
        out, cache = self.model(inputs, cache, inputs_embeds)
        return self.lm_head(out), cache

    @staticmethod
    def sanitize(weights):