Method build_input_k_idxs

subprojects/llama.cpp/src/llama-kv-cache.cpp:1159–1167

}

ggml_tensor * llama_kv_cache::build_input_k_idxs(ggml_context * ctx, const llama_ubatch & ubatch) const {
    const uint32_t n_tokens = ubatch.n_tokens;

    ggml_tensor * k_idxs = ggml_new_tensor_1d(ctx, GGML_TYPE_I64, n_tokens);

    ggml_set_input(k_idxs);

    return k_idxs;
}

ggml_tensor * llama_kv_cache::build_input_v_idxs(ggml_context * ctx, const llama_ubatch & ubatch) const {
    const uint32_t n_tokens = ubatch.n_tokens;