Method llm_build_glm4_moe

subprojects/llama.cpp/src/models/glm4-moe.cpp:3–170

#include "models.h"

llm_build_glm4_moe::llm_build_glm4_moe(const llama_model & model, const llm_graph_params & params) : llm_graph_context(params) {
    const int64_t n_embd_head = hparams.n_embd_head_v;

    GGML_ASSERT(n_embd_head == hparams.n_embd_head_k);

    int sections[4];
    std::copy(std::begin(hparams.rope_sections), std::begin(hparams.rope_sections) + 4, sections);

    ggml_tensor * cur;
    ggml_tensor * inpL;

    inpL = build_inp_embd(model.tok_embd);

    bool use_mrope = hparams.use_mrope();
    if (ubatch.embd && !use_mrope) {
        // unfortunately, we need to forcefully stop here, to avoid users complaining about wrong results
        GGML_ABORT("This GGUF does not support multimodal. Please reconvert it.");
    }

    // inp_pos - contains the positions
    ggml_tensor * inp_pos = build_inp_pos();

    auto * inp_attn = build_attn_inp_kv();

    ggml_tensor * inp_out_ids = build_inp_out_ids();

    // Only process up to last layer (skip final NextN layer)
    // Final layer tensors are loaded but not processed in forward pass
    const int n_transformer_layers = n_layer - hparams.nextn_predict_layers;
    for (int il = 0; il < n_transformer_layers; ++il) {
        ggml_tensor * inpSA = inpL;

        // Pre-attention norm
        cur = build_norm(inpL, model.layers[il].attn_norm, NULL, LLM_NORM_RMS, il);
        cb(cur, "attn_norm", il);

        // self-attention
        {
            ggml_tensor * Qcur = build_lora_mm(model.layers[il].wq, cur);
            if (model.layers[il].bq) {
                Qcur = ggml_add(ctx0, Qcur, model.layers[il].bq);
            }
            cb(Qcur, "Qcur", il);

            ggml_tensor * Kcur = build_lora_mm(model.layers[il].wk, cur);
            if (model.layers[il].bk) {
                Kcur = ggml_add(ctx0, Kcur, model.layers[il].bk);
            }
            cb(Kcur, "Kcur", il);

            ggml_tensor * Vcur = build_lora_mm(model.layers[il].wv, cur);
            if (model.layers[il].bv) {
                Vcur = ggml_add(ctx0, Vcur, model.layers[il].bv);
            }
            cb(Vcur, "Vcur", il);

            Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens);
            Kcur = ggml_reshape_3d(ctx0, Kcur, n_embd_head, n_head_kv, n_tokens);