Function common_sampler_sample

subprojects/llama.cpp/common/sampling.cpp:449–519

}

llama_token common_sampler_sample(struct common_sampler * gsmpl, struct llama_context * ctx, int idx, bool grammar_first) {
    llama_synchronize(ctx);

    // start measuring sampling time after the llama_context synchronization in order to not measure any ongoing async operations
    const auto tm = gsmpl->tm();

    llama_token id = LLAMA_TOKEN_NULL;

    auto & grmr  = gsmpl->grmr;
    auto & chain = gsmpl->chain;
    auto & cur_p = gsmpl->cur_p; // initialized by set_logits

    // Check if a backend sampler has already sampled a token in which case we
    // return that token id directly.
    {
        id = llama_get_sampled_token_ith(ctx, idx);

        if (id != LLAMA_TOKEN_NULL) {
            LOG_DBG("%s: Backend sampler selected token: '%d'. Will not run any CPU samplers\n", __func__, id);

            GGML_ASSERT(!gsmpl->grmr && "using grammar in combination with backend sampling is not supported");

            // TODO: simplify
            gsmpl->cur.resize(1);
            gsmpl->cur[0] = { id, 0.0f, 1.0f };
            cur_p = { gsmpl->cur.data(), gsmpl->cur.size(), 0, true };

            return id;
        }
    }

    gsmpl->set_logits(ctx, idx);

    if (grammar_first) {
        llama_sampler_apply(grmr, &cur_p);
    }

    llama_sampler_apply(chain, &cur_p);

    id = cur_p.data[cur_p.selected].id;

    if (grammar_first) {
        return id;
    }

    // check if it the sampled token fits the grammar (grammar-based rejection sampling)
    {
        llama_token_data       single_token_data       = { id, 1.0f, 0.0f };
        llama_token_data_array single_token_data_array = { &single_token_data, 1, -1, false };

        llama_sampler_apply(grmr, &single_token_data_array);

        const bool is_valid = single_token_data_array.data[0].logit != -INFINITY;
        if (is_valid) {
            return id;
        }
    }