echodict/llama.cpp

1#include "models.h"2 3template <bool iswa>4llm_build_olmo2<iswa>::llm_build_olmo2(const llama_model & model, const llm_graph_params & params) : llm_graph_context(params) {5    const int64_t n_embd_head = hparams.n_embd_head_v();6 7    GGML_ASSERT(n_embd_head == hparams.n_embd_head_k());8    GGML_ASSERT(n_embd_head == n_rot);9 10    ggml_tensor * cur;11    ggml_tensor * inpL;12 13    inpL = build_inp_embd(model.tok_embd);14 15    // inp_pos - contains the positions16    ggml_tensor * inp_pos = build_inp_pos();17 18    using inp_attn_type = std::conditional_t<iswa, llm_graph_input_attn_kv_iswa, llm_graph_input_attn_kv>;19    inp_attn_type * inp_attn = nullptr;20 21    if constexpr (iswa) {22        inp_attn = build_attn_inp_kv_iswa();23    } else {24        inp_attn = build_attn_inp_kv();25    }26    ggml_tensor * inp_out_ids = build_inp_out_ids();27 28    for (int il = 0; il < n_layer; ++il) {29        ggml_tensor * inpSA = inpL;30 31        cur = inpL;32 33        // self_attention34        {35            // compute Q and K and RoPE them36            ggml_tensor * Qcur = build_lora_mm(model.layers[il].wq, cur);37            cb(Qcur, "Qcur", il);38 39            ggml_tensor * Kcur = build_lora_mm(model.layers[il].wk, cur);40            cb(Kcur, "Kcur", il);41 42            ggml_tensor * Vcur = build_lora_mm(model.layers[il].wv, cur);43            cb(Vcur, "Vcur", il);44 45            Qcur = build_norm(Qcur, model.layers[il].attn_q_norm, NULL,46                    LLM_NORM_RMS, il);47            cb(Qcur, "Qcur_normed", il);48 49            Kcur = build_norm(Kcur, model.layers[il].attn_k_norm, NULL,50                    LLM_NORM_RMS, il);51            cb(Kcur, "Kcur_normed", il);52 53            Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head,    n_tokens);54            Kcur = ggml_reshape_3d(ctx0, Kcur, n_embd_head, n_head_kv, n_tokens);55            Vcur = ggml_reshape_3d(ctx0, Vcur, n_embd_head, n_head_kv, n_tokens);56 57            const bool is_swa = hparams.is_swa(il);58 59            if (is_swa) {60                // For sliding window layers, Olmo3 use regular rope with no yarn rope scaling.61                // This is achieved here by setting freq_scale and attn_factor to 1.62                // We also set ext_factor to 0 to avoid a few unnecessary computations.63                Qcur = ggml_rope_ext(64                    ctx0, Qcur, inp_pos, nullptr,65                    n_rot, rope_type, n_ctx_orig, freq_base, 1.0,66                    0.0, 1.0, beta_fast, beta_slow67                    );68 69                Kcur = ggml_rope_ext(70                    ctx0, Kcur, inp_pos, nullptr,71                    n_rot, rope_type, n_ctx_orig, freq_base, 1.0,72                    0.0, 1.0, beta_fast, beta_slow73                    );74            } else {75                Qcur = ggml_rope_ext(76                    ctx0, Qcur, inp_pos, nullptr,77                    n_rot, rope_type, n_ctx_orig, freq_base, freq_scale,78                    ext_factor, attn_factor, beta_fast, beta_slow79                    );80 81                Kcur = ggml_rope_ext(82                    ctx0, Kcur, inp_pos, nullptr,83                    n_rot, rope_type, n_ctx_orig, freq_base, freq_scale,84                    ext_factor, attn_factor, beta_fast, beta_slow85                    );86            }87            cb(Qcur, "Qcur", il);88            cb(Kcur, "Kcur", il);89            cb(Vcur, "Vcur", il);90 91            cur = build_attn(inp_attn,92                    model.layers[il].wo, NULL, model.layers[il].wo_s,93                    Qcur, Kcur, Vcur, nullptr, nullptr, nullptr, 1.0f/sqrtf(float(n_embd_head)), il);94        }95        if (il == n_layer - 1 && inp_out_ids) {96            cur   = ggml_get_rows(ctx0,   cur, inp_out_ids);97            inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids);98        }99        cur = build_norm(cur,100                model.layers[il].attn_post_norm, NULL,101                LLM_NORM_RMS, il);102        cb(cur, "attn_post_norm", il);103 104        ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA);105        cb(ffn_inp, "ffn_inp", il);106 107        // feed-forward network108        cur = build_ffn(ffn_inp,109                model.layers[il].ffn_up,   NULL, NULL,110                model.layers[il].ffn_gate, NULL, NULL,111                model.layers[il].ffn_down, NULL, NULL,112                NULL,113                LLM_FFN_SILU, LLM_FFN_PAR, il);114        cb(cur, "ffn_out", il);115 116        cur = build_norm(cur,117                model.layers[il].ffn_post_norm, NULL,118                LLM_NORM_RMS, -1);119        cb(cur, "ffn_post_norm", -1);120 121        cur = ggml_add(ctx0, cur, ffn_inp);122        cb(cur, "ffn_out", il);123 124        cur = build_cvec(cur, il);125        cb(cur, "l_out", il);126 127        // input for next layer128        inpL = cur;129    }130    cur = inpL;131 132    cur = build_norm(cur,133            model.output_norm, NULL,134            LLM_NORM_RMS, -1);135 136    cb(cur, "result_norm", -1);137    res->t_embd = cur;138 139    // lm_head140    cur = build_lora_mm(model.output, cur);141 142    cb(cur, "result_output", -1);143    res->t_logits = cur;144 145    ggml_build_forward_expand(gf, cur);146}147 148// Explicit template instantiations149template struct llm_build_olmo2<false>;150template struct llm_build_olmo2<true>;151