File size: 11,240 Bytes
0475af5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
#include "models.h"

void llama_model_exaone_moe::load_arch_hparams(llama_model_loader & ml) {
    hparams.swa_type = LLAMA_SWA_TYPE_STANDARD;
    hparams.n_swa = 128;
    uint32_t swa_period = 4;
    ml.get_key_or_arr(LLM_KV_ATTENTION_SLIDING_WINDOW_PATTERN, swa_period, false);
    hparams.set_swa_pattern(swa_period);
    hparams.rope_freq_base_train_swa  = hparams.rope_freq_base_train;
    hparams.rope_freq_scale_train_swa = hparams.rope_freq_scale_train;

    ml.get_key(LLM_KV_ROPE_FREQ_BASE_SWA,                hparams.rope_freq_base_train_swa, false);
    ml.get_key(LLM_KV_ATTENTION_SLIDING_WINDOW,          hparams.n_swa);
    ml.get_key(LLM_KV_ATTENTION_LAYERNORM_RMS_EPS,       hparams.f_norm_rms_eps);
    ml.get_key(LLM_KV_EXPERT_SHARED_COUNT,               hparams.n_expert_shared, false);
    ml.get_key(LLM_KV_EXPERT_FEED_FORWARD_LENGTH,        hparams.n_ff_exp);
    ml.get_key(LLM_KV_EXPERT_SHARED_FEED_FORWARD_LENGTH, hparams.n_ff_shexp, false);
    ml.get_key(LLM_KV_EXPERT_GATING_FUNC,                hparams.expert_gating_func);
    ml.get_key(LLM_KV_EXPERT_WEIGHTS_SCALE,              hparams.expert_weights_scale, false);
    ml.get_key(LLM_KV_EXPERT_WEIGHTS_NORM,               hparams.expert_weights_norm, false);
    ml.get_key(LLM_KV_LEADING_DENSE_BLOCK_COUNT,         hparams.n_layer_dense_lead, false);

    ml.get_key(LLM_KV_NEXTN_PREDICT_LAYERS, hparams.n_layer_nextn, false);
    GGML_ASSERT(hparams.n_layer_nextn < hparams.n_layer_all && "n_layer_nextn must be < n_layer_impl");

    switch (hparams.n_layer()) {
        case 32: type = LLM_TYPE_30B_A3B; break;
        case 48: type = LLM_TYPE_235B_A22B; break;
        default: type = LLM_TYPE_UNKNOWN;
    }
}

void llama_model_exaone_moe::load_arch_tensors(llama_model_loader &) {
    LLAMA_LOAD_LOCALS;

    const int64_t n_ff_exp       = hparams.n_ff_exp;
    const int64_t n_ff_shexp     = hparams.n_ff_shexp > 0 ? hparams.n_ff_shexp : n_ff_exp;
    const int64_t head_dim       = hparams.n_embd_head_k();
    const int64_t n_qo_dim       = n_head * head_dim;
    const int64_t n_kv_dim       = n_head_kv * head_dim;

    tok_embd = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, 0);

    // output
    output_norm = create_tensor(tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd}, 0);
    output      = create_tensor(tn(LLM_TENSOR_OUTPUT,      "weight"), {n_embd, n_vocab}, 0);

    if (output == NULL) {
        output = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, TENSOR_DUPLICATED);
    }

    for (int i = 0; i < n_layer_all; ++i) {
        int flags = 0;
        if (i >= n_layer) {
            // skip all tensors in the NextN layers
            flags |= TENSOR_SKIP;
        }

        auto & layer = layers[i];
        create_tensor_qkv(layer, i, n_embd, n_qo_dim, n_kv_dim, n_kv_dim, flags);
        layer.wo = create_tensor(tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_qo_dim, n_embd}, flags);

        layer.rope_freqs   = create_tensor(tn(LLM_TENSOR_ROPE_FREQS,  "weight", i), {n_rot/2}, TENSOR_NOT_REQUIRED | (i != 0 ? TENSOR_DUPLICATED : 0) | flags);

        layer.attn_norm    = create_tensor(tn(LLM_TENSOR_ATTN_NORM,   "weight", i), {n_embd}, flags);
        layer.attn_q_norm  = create_tensor(tn(LLM_TENSOR_ATTN_Q_NORM, "weight", i), {n_embd_head_k}, flags);
        layer.attn_k_norm  = create_tensor(tn(LLM_TENSOR_ATTN_K_NORM, "weight", i), {n_embd_head_k}, flags);

        layer.ffn_norm     = create_tensor(tn(LLM_TENSOR_FFN_NORM,    "weight", i), {n_embd}, flags);

        // dense layers for first n_layer_dense_lead layers or nextn_predict_layers layers at the end
        if (i < (int) hparams.n_layer_dense_lead || (i >= n_layer)) {
            layer.ffn_gate = create_tensor(tn(LLM_TENSOR_FFN_GATE, "weight", i), {n_embd, n_ff}, flags);
            layer.ffn_down = create_tensor(tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, flags);
            layer.ffn_up   = create_tensor(tn(LLM_TENSOR_FFN_UP,   "weight", i), {n_embd, n_ff}, flags);
        } else {
            layer.ffn_gate_inp    = create_tensor(tn(LLM_TENSOR_FFN_GATE_INP,  "weight", i), {n_embd, n_expert}, flags);
            layer.ffn_exp_probs_b = create_tensor(tn(LLM_TENSOR_FFN_EXP_PROBS_B, "bias", i), {n_expert}, TENSOR_NOT_REQUIRED | flags);

            if (n_expert == 0) {
                throw std::runtime_error("n_expert must be > 0");
            }
            if (n_expert_used == 0) {
                throw std::runtime_error("n_expert_used must be > 0");
            }

            layer.ffn_gate_exps  = create_tensor(tn(LLM_TENSOR_FFN_GATE_EXPS,  "weight", i), {n_embd, n_ff_exp, n_expert}, flags);
            layer.ffn_down_exps  = create_tensor(tn(LLM_TENSOR_FFN_DOWN_EXPS,  "weight", i), {n_ff_exp, n_embd, n_expert}, flags);
            layer.ffn_up_exps    = create_tensor(tn(LLM_TENSOR_FFN_UP_EXPS,    "weight", i), {n_embd, n_ff_exp, n_expert}, flags);

            layer.ffn_gate_shexp = create_tensor(tn(LLM_TENSOR_FFN_GATE_SHEXP, "weight", i), {n_embd, n_ff_shexp}, flags);
            layer.ffn_down_shexp = create_tensor(tn(LLM_TENSOR_FFN_DOWN_SHEXP, "weight", i), {n_ff_shexp, n_embd}, flags);
            layer.ffn_up_shexp   = create_tensor(tn(LLM_TENSOR_FFN_UP_SHEXP,   "weight", i), {n_embd, n_ff_shexp}, flags);
        }

        // NextN/MTP tensors (preserved but unused) - conditionally load for last nextn_predict_layers
        if (i >= n_layer) {
            layer.nextn.eh_proj          = create_tensor(tn(LLM_TENSOR_NEXTN_EH_PROJ, "weight", i), {2 * n_embd, n_embd}, flags);
            layer.nextn.enorm            = create_tensor(tn(LLM_TENSOR_NEXTN_ENORM,   "weight", i), {n_embd}, flags);
            layer.nextn.hnorm            = create_tensor(tn(LLM_TENSOR_NEXTN_HNORM,   "weight", i), {n_embd}, flags);

            layer.nextn.shared_head_norm = create_tensor(tn(LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, "weight", i), {n_embd}, flags | TENSOR_NOT_REQUIRED);
            layer.nextn.embed_tokens     = create_tensor(tn(LLM_TENSOR_NEXTN_EMBED_TOKENS,     "weight", i), {n_embd, n_vocab}, flags | TENSOR_NOT_REQUIRED);
            layer.nextn.shared_head_head = create_tensor(tn(LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, "weight", i), {n_embd, n_vocab}, flags | TENSOR_NOT_REQUIRED);
        }
    }
}

std::unique_ptr<llm_graph_context> llama_model_exaone_moe::build_arch_graph(const llm_graph_params & params) const {
    return std::make_unique<graph>(*this, params);
}

llama_model_exaone_moe::graph::graph(const llama_model & model, const llm_graph_params & params) :
    llm_graph_context(params) {
    const int64_t n_embd_head = hparams.n_embd_head_k();

    GGML_ASSERT(n_embd_head == hparams.n_embd_head_v());
    GGML_ASSERT(n_embd_head == n_rot);

    ggml_tensor * cur;
    ggml_tensor * inpL;

    inpL = build_inp_embd(model.tok_embd);

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

    auto * inp_attn_iswa = build_attn_inp_kv_iswa();

    ggml_tensor * inp_out_ids = build_inp_out_ids();

    for (int il = 0; il < n_layer; ++il) {
        ggml_tensor * inpSA = inpL;

        // use RoPE for SWA layers
        const bool is_local_layer = hparams.is_swa(il);

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

        // self-attention
        {
            ggml_tensor * rope_factors = model.get_rope_factors(cparams, il);

            // compute Q and K and RoPE them
            auto [Qcur, Kcur, Vcur] = build_qkv(model.layers[il], cur,
                    n_embd_head, n_head, n_head_kv, il);

            Qcur = build_norm(Qcur, model.layers[il].attn_q_norm, NULL, LLM_NORM_RMS, il);
            Kcur = build_norm(Kcur, model.layers[il].attn_k_norm, NULL, LLM_NORM_RMS, il);
            cb(Qcur, "Qcur_normed", il);
            cb(Kcur, "Kcur_normed", il);

            if (is_local_layer) {
                Qcur = ggml_rope_ext(ctx0, Qcur, inp_pos, rope_factors, n_rot, rope_type, n_ctx_orig, freq_base,
                                     freq_scale, ext_factor, attn_factor, beta_fast, beta_slow);

                Kcur = ggml_rope_ext(ctx0, Kcur, inp_pos, rope_factors, n_rot, rope_type, n_ctx_orig, freq_base,
                                     freq_scale, ext_factor, attn_factor, beta_fast, beta_slow);
            }
            cb(Qcur, "Qcur", il);
            cb(Kcur, "Kcur", il);
            cb(Vcur, "Vcur", il);

            cur = build_attn(inp_attn_iswa,
                model.layers[il].wo, NULL, model.layers[il].wo_s,
                Qcur, Kcur, Vcur, nullptr, nullptr, nullptr, 1.0f / sqrtf(float(n_embd_head)), il);
            cb(cur, "attn_out", il);
        }
        if (il == n_layer - 1 && inp_out_ids) {
            cur   = ggml_get_rows(ctx0, cur, inp_out_ids);
            inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids);
        }
        ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA);
        cb(ffn_inp, "ffn_inp", il);

        // norm
        cur = build_norm(ffn_inp, model.layers[il].ffn_norm, NULL, LLM_NORM_RMS, il);
        cb(cur, "ffn_norm", il);

        // feed-forward network
        if (model.layers[il].ffn_gate_inp == nullptr) {
            // dense branch
            cur = build_ffn(cur,
                    model.layers[il].ffn_up, NULL, NULL,
                    model.layers[il].ffn_gate, NULL, NULL,
                    model.layers[il].ffn_down, NULL, NULL, NULL,
                    LLM_FFN_SILU, LLM_FFN_PAR, il);
            cb(cur, "ffn_out", il);
        } else {
            // MoE branch
            ggml_tensor * moe_out = build_moe_ffn(cur,
                model.layers[il].ffn_gate_inp,
                model.layers[il].ffn_up_exps,
                model.layers[il].ffn_gate_exps,
                model.layers[il].ffn_down_exps,
                model.layers[il].ffn_exp_probs_b,
                n_expert, n_expert_used,
                LLM_FFN_SILU, hparams.expert_weights_norm,
                hparams.expert_weights_scale,
                (llama_expert_gating_func_type) hparams.expert_gating_func,
                il);
            cb(moe_out, "ffn_moe_out", il);

            // FFN shared expert
            {
                ggml_tensor * ffn_shexp =
                    build_ffn(cur,
                        model.layers[il].ffn_up_shexp, NULL, NULL,
                        model.layers[il].ffn_gate_shexp, NULL, NULL,
                        model.layers[il].ffn_down_shexp, NULL, NULL,
                        NULL, LLM_FFN_SILU, LLM_FFN_PAR, il);
                cb(ffn_shexp, "ffn_shexp", il);

                cur = ggml_add(ctx0, moe_out, ffn_shexp);
                cb(cur, "ffn_out", il);
            }
        }

        cur = ggml_add(ctx0, cur, ffn_inp);

        cur = build_cvec(cur, il);
        cb(cur, "l_out", il);

        // input for next layer
        inpL = cur;
    }
    cur = inpL;

    // final norm
    cur = build_norm(cur, model.output_norm, NULL, LLM_NORM_RMS, -1);

    cb(cur, "result_norm", -1);
    res->t_embd = cur;

    // lm_head
    cur = build_lora_mm(model.output, cur, model.output_s);

    cb(cur, "result_output", -1);
    res->t_logits = cur;

    ggml_build_forward_expand(gf, cur);
}