Method __init__

inference/models/llama.py:304–308

(self, params)

class LlamaForCausalLM(nn.Module):
    def __init__(self, params):
        super().__init__()
        self.params = params
        self.model = Transformer(params)
        self.lm_head = nn.Linear(params.hidden_size, params.vocab_size, bias=False)

    @torch.inference_mode()
    def forward(self, tokens: torch.Tensor, start_pos: int):