wmantly/llama.cpp-xrip
1
0 Fork 0
Code Issues Pull Requests Actions 33 Packages Projects Releases Wiki Activity

model: mistral le chonk (#31376)

Browse Source 
* llama: extend DSA and sdpark to new model

* fix ci
This commit is contained in:
Xuan-Son Nguyen authored and GitHub committed 2026-10-29 15:43:33 +01:00
1 parent 0ac1a7fd35
commit 78feec062d
23 files changed
+1290 -35
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

src/models/dflash.cpp
+328 -17
View File
@@ -112,6 +112,34 @@ void llama_model_dflash::load_arch_hparams(llama_model_loader & ml) {
return;
}
// Mistral Large 4 DSpark: dense MLA stages, one sliding window, plain YaRN
if (ml.get_key(LLM_KV_ATTENTION_KV_LORA_RANK, hparams.n_lora_kv, false)) {
ml.get_key(LLM_KV_ATTENTION_KEY_LENGTH_MLA, hparams.n_embd_head_k_mla_impl);
ml.get_key(LLM_KV_ATTENTION_VALUE_LENGTH_MLA, hparams.n_embd_head_v_mla_impl);
ml.get_key(LLM_KV_ROPE_SCALING_YARN_LOG_MUL, hparams.rope_yarn_log_mul, false);
ml.get_key(LLM_KV_ATTENTION_SLIDING_WINDOW, hparams.n_swa);
if (hparams.n_swa == 0) {
throw std::runtime_error("DSpark Mistral Large 4 draft expects a sliding window");
}
if (hparams.n_embd_head_k_full != hparams.n_lora_kv + hparams.n_rot()) {
throw std::runtime_error("DSpark Mistral Large 4 draft expects key_length == kv_lora_rank + rope dim");
}
// same as DSV4.1 DSpark: block uses the window of the anchor
hparams.n_swa += 1;
hparams.swa_type = LLAMA_SWA_TYPE_STANDARD;
hparams.set_swa_pattern(0);
for (uint32_t il = 0; il < hparams.n_layer_all; ++il) {
hparams.is_swa_impl[il] = true;
}
hparams.rope_freq_base_train_swa = hparams.rope_freq_base_train;
hparams.rope_freq_scale_train_swa = hparams.rope_freq_scale_train;
type = LLM_TYPE_UNKNOWN;
return;
}
// optional interleaved sliding-window attention with per-layer pattern array.
// DFlash has a single rope, so the SWA rope == main rope.
if (ml.get_key(LLM_KV_ATTENTION_SLIDING_WINDOW, hparams.n_swa, false) && hparams.n_swa > 0) {
@@ -152,6 +180,9 @@ void llama_model_dflash::load_arch_tensors(llama_model_loader &) {
dspark_conf_proj = create_tensor(tn(LLM_TENSOR_DSPARK_CONF_PROJ, "weight"), { n_embd + dspark_markov_rank, 1 }, TENSOR_NOT_REQUIRED);
dspark_conf_proj_b = create_tensor(tn(LLM_TENSOR_DSPARK_CONF_PROJ, "bias"), { 1 }, TENSOR_NOT_REQUIRED);
dspark_joint_proj = create_tensor(tn(LLM_TENSOR_DSPARK_JOINT_PROJ, "weight"), { n_embd + 2*dspark_markov_rank, 3*dspark_markov_rank }, TENSOR_NOT_REQUIRED);
dspark_joint_proj_b = create_tensor(tn(LLM_TENSOR_DSPARK_JOINT_PROJ, "bias"), { 3*dspark_markov_rank }, dspark_joint_proj ? 0 : TENSOR_NOT_REQUIRED);
LLAMA_LOG_INFO("%s: DFlash with DSpark markov head (rank = %lld)\n", __func__, (long long) dspark_markov_rank);
}
@@ -252,6 +283,33 @@ void llama_model_dflash::load_arch_tensors(llama_model_loader &) {
// a draft with its own embeddings + head references no target tensors and can run on devices the target does not use (e.g. -devd with a tensor-split target)
output = create_tensor(tn(LLM_TENSOR_OUTPUT, "weight"), { n_embd, n_vocab_draft }, TENSOR_NOT_REQUIRED);
if (hparams.n_lora_kv > 0) {
const int64_t kv_lora_rank = hparams.n_lora_kv;
const int64_t n_embd_head_k_mla = hparams.n_embd_head_k_mla();
const int64_t n_embd_head_v_mla = hparams.n_embd_head_v_mla();
const int64_t n_embd_head_nope = n_embd_head_k_mla - n_rot;
dspark_mask_embd = create_tensor(tn(LLM_TENSOR_DSPARK_MASK_EMBD, "weight"), { n_embd, hparams.dflash_block_size - 1 }, 0);
for (int i = 0; i < n_layer; ++i) {
auto & layer = layers[i];
layer.attn_norm = create_tensor(tn(LLM_TENSOR_ATTN_NORM, "weight", i), { n_embd }, 0);
layer.wq = create_tensor(tn(LLM_TENSOR_ATTN_Q, "weight", i), { n_embd, n_head * n_embd_head_k_mla }, 0);
layer.wkv_a_mqa = create_tensor(tn(LLM_TENSOR_ATTN_KV_A_MQA, "weight", i), { n_embd, kv_lora_rank + n_rot }, 0);
layer.attn_kv_a_norm = create_tensor(tn(LLM_TENSOR_ATTN_KV_A_NORM, "weight", i), { kv_lora_rank }, 0);
layer.wk_b = create_tensor(tn(LLM_TENSOR_ATTN_K_B, "weight", i), { n_embd_head_nope, kv_lora_rank, n_head }, 0);
layer.wv_b = create_tensor(tn(LLM_TENSOR_ATTN_V_B, "weight", i), { kv_lora_rank, n_embd_head_v_mla, n_head }, 0);
layer.wo = create_tensor(tn(LLM_TENSOR_ATTN_OUT, "weight", i), { n_head * n_embd_head_v_mla, n_embd }, 0);
layer.ffn_norm = create_tensor(tn(LLM_TENSOR_FFN_NORM, "weight", i), { n_embd }, 0);
layer.ffn_gate = create_tensor(tn(LLM_TENSOR_FFN_GATE, "weight", i), { n_embd, n_ff }, 0);
layer.ffn_down = create_tensor(tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd }, 0);
layer.ffn_up = create_tensor(tn(LLM_TENSOR_FFN_UP, "weight", i), { n_embd, n_ff }, 0);
}
return;
}
if (hparams.dsv4_hc_mult > 0) {
const int64_t q_lora_rank = hparams.n_lora_q;
const int64_t n_ff_exp = hparams.n_ff_exp();
@@ -377,7 +435,7 @@ llama_model_dflash::graph<true>::graph(const llama_model & model, const llm_grap
}
// DSpark (DFlash + Markov & Confidence head): Markov bias on the draft logits, chained per block position
static void build_dspark_markov_head(llm_graph_context & g, const llama_model & model, ggml_tensor * tokens) {
static void build_dspark_markov_head(llm_graph_context & g, const llama_model & model, ggml_tensor * tokens, ggml_tensor * hidden = nullptr) {
ggml_context * ctx0 = g.ctx0;
auto & res = g.res;
@@ -388,6 +446,10 @@ static void build_dspark_markov_head(llm_graph_context & g, const llama_model &
// confidence head is optional
const bool has_conf = model.dspark_conf_proj != nullptr;
// rnn head: gated state carried across block positions
const bool is_rnn = model.dspark_joint_proj != nullptr;
GGML_ASSERT(!is_rnn || hidden != nullptr);
ggml_tensor * base = res->t_logits; // [n_vocab, n_tokens]
const int64_t n_vocab = base->ne[0];
const int64_t n_tok = base->ne[1];
@@ -428,11 +490,37 @@ static void build_dspark_markov_head(llm_graph_context & g, const llama_model &
}
}
ggml_tensor * state = nullptr; // [R, n_blocks]
// TODO: the in-graph chain is greedy (argmax); sampling params affect only the final
// token pick, not the Markov conditioning path
for (int64_t i = i_draft_beg; i < block_drafts; ++i) {
ggml_tensor * w1_prev = ggml_get_rows(ctx0, w1, prev); // [R, n_blocks]
ggml_tensor * bias = g.build_lora_mm(w2, w1_prev, model.dspark_markov_w2_s); // [n_vocab_draft, n_blocks]
ggml_tensor * w1_prev = ggml_get_rows(ctx0, w1, prev); // [R, n_blocks]
ggml_tensor * markov = w1_prev;
if (is_rnn) {
const int64_t rank = w1_prev->ne[0];
if (state == nullptr) {
state = ggml_fill(ctx0, ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, rank, n_blocks), 0.0f);
}
ggml_tensor * hidden_i = ggml_view_2d(ctx0, hidden, hidden->ne[0], n_blocks,
(size_t) block_drafts * hidden->nb[1], i*hidden->nb[1]);
ggml_tensor * joint = ggml_concat(ctx0, ggml_concat(ctx0, state, w1_prev, 0), ggml_cont(ctx0, hidden_i), 0);
joint = ggml_add(ctx0, ggml_mul_mat(ctx0, model.dspark_joint_proj, joint), model.dspark_joint_proj_b); // [3R, n_blocks]
const auto chunk = [&](int64_t k) {
return ggml_cont(ctx0, ggml_view_2d(ctx0, joint, rank, n_blocks, joint->nb[1], k*rank*ggml_element_size(joint)));
};
ggml_tensor * gate = ggml_sigmoid(ctx0, chunk(0));
ggml_tensor * cand = ggml_tanh(ctx0, chunk(1));
// state = gate * state + (1 - gate) * cand
state = ggml_add(ctx0, cand, ggml_mul(ctx0, gate, ggml_sub(ctx0, state, cand)));
markov = ggml_tanh(ctx0, chunk(2));
}
ggml_tensor * bias = g.build_lora_mm(w2, markov, model.dspark_markov_w2_s); // [n_vocab_draft, n_blocks]
if (model.d2t) {
// reduced draft vocab: scatter the bias to the target rows (base is -inf on the others)
const int64_t n_draft_vocab = bias->ne[0];
@@ -455,7 +543,7 @@ static void build_dspark_markov_head(llm_graph_context & g, const llama_model &
// conf(i) = sigmoid(conf_proj . [conf_inp(i); markov_w1[prev(i)]] + b) -- [1, n_blocks]
ggml_tensor * conf_inp_i = ggml_view_2d(ctx0, conf_inp, conf_inp->ne[0], n_blocks,
(size_t) block_drafts * conf_inp->nb[1], i*conf_inp->nb[1]);
ggml_tensor * feat = ggml_concat(ctx0, ggml_cont(ctx0, conf_inp_i), w1_prev, 0);
ggml_tensor * feat = ggml_concat(ctx0, ggml_cont(ctx0, conf_inp_i), markov, 0);
ggml_tensor * conf = ggml_mul_mat(ctx0, model.dspark_conf_proj, feat);
if (model.dspark_conf_proj_b) {
conf = ggml_add(ctx0, conf, model.dspark_conf_proj_b);
@@ -1291,6 +1379,23 @@ llama_model_dflash::graph_dsv4::graph_dsv4(const llama_model & model, const llm_
// DeepSeek-V4.1 DSpark decoder (mtp.0..2), dual-mode by batch type like graph_dsv4:
// * embd batch -> writes every stage's kv_norm(wkv(main_proj(target features))) into its window ring
// * token batch -> [anchor, noise...] block through the stages, single-pass mHC, no per-head q norm
// window of a block = window of its first query, see n_swa
static void build_dspark_block_mask(llm_graph_context & g, llm_graph_input_attn_k_iswa * inp_attn) {
ggml_context * ctx0 = g.ctx0;
ggml_tensor * m = inp_attn->self_kq_mask_swa_cnv;
const int64_t n_tokens = g.n_tokens;
const int64_t n_blocks = g.ubatch.n_seqs_unq;
const int64_t bd = n_blocks > 0 ? n_tokens / n_blocks : 0;
if (m && bd > 1 && n_tokens % n_blocks == 0 && m->ne[2] == 1 && m->ne[1] % bd == 0 && ggml_is_contiguous(m)) {
ggml_tensor * m4 = ggml_reshape_4d(ctx0, m, m->ne[0], bd, m->ne[1]/bd, m->ne[3]);
ggml_tensor * r0 = ggml_view_4d(ctx0, m4, m4->ne[0], 1, m4->ne[2], m4->ne[3], m4->nb[1], m4->nb[2], m4->nb[3], 0);
m4 = ggml_repeat(ctx0, r0, m4);
inp_attn->self_kq_mask_swa_cnv = ggml_reshape_4d(ctx0, m4, m->ne[0], m->ne[1], 1, m->ne[3]);
g.cb(inp_attn->self_kq_mask_swa_cnv, "dspark_block_mask", -1);
}
}
llama_model_dflash::graph_dsv41::graph_dsv41(const llama_model & model, const llm_graph_params & params) :
llama_model_deepseek41::graph(params) {
const int64_t n_embd_inp = hparams.n_embd_inp_enc();
@@ -1382,19 +1487,7 @@ llama_model_dflash::graph_dsv41::graph_dsv41(const llama_model & model, const ll
inpL = ggml_repeat_4d(ctx0, inpL, n_embd, hc, n_tokens, 1);
cb(inpL, "hc_init", -1);
// window of a block = window of its first query, see n_swa
{
ggml_tensor * m = inp_attn->self_kq_mask_swa_cnv;
const int64_t n_blocks = ubatch.n_seqs_unq;
const int64_t bd = n_blocks > 0 ? n_tokens / n_blocks : 0;
if (m && bd > 1 && n_tokens % n_blocks == 0 && m->ne[2] == 1 && m->ne[1] % bd == 0 && ggml_is_contiguous(m)) {
ggml_tensor * m4 = ggml_reshape_4d(ctx0, m, m->ne[0], bd, m->ne[1]/bd, m->ne[3]);
ggml_tensor * r0 = ggml_view_4d(ctx0, m4, m4->ne[0], 1, m4->ne[2], m4->ne[3], m4->nb[1], m4->nb[2], m4->nb[3], 0);
m4 = ggml_repeat(ctx0, r0, m4);
inp_attn->self_kq_mask_swa_cnv = ggml_reshape_4d(ctx0, m4, m->ne[0], m->ne[1], 1, m->ne[3]);
cb(inp_attn->self_kq_mask_swa_cnv, "dspark_block_mask", -1);
}
}
build_dspark_block_mask(*this, inp_attn);
ggml_tensor * pre_prev = nullptr;
@@ -1529,12 +1622,230 @@ llama_model_dflash::graph_dsv41::graph_dsv41(const llama_model & model, const ll
}
}
// Mistral Large 4 DSpark, dual-mode like graph_dsv41:
// - embd batch: write latent K of target_norm(fc(features)) to the cache, no attn norm
// - token batch: decode [anchor, mask_1, ...] block, non-causal
llama_model_dflash::graph_mistral_large4::graph_mistral_large4(const llama_model & model, const llm_graph_params & params) : llm_graph_context(params) {
const int64_t n_embd_inp = hparams.n_embd_inp_enc();
const int64_t kv_lora_rank = hparams.n_lora_kv;
const int64_t n_embd_head_k_mla = hparams.n_embd_head_k_mla();
const int64_t n_embd_head_v_mla = hparams.n_embd_head_v_mla();
const int64_t n_embd_head_nope = n_embd_head_k_mla - n_rot;
const float kq_scale = 1.0f/sqrtf(float(n_embd_head_k_mla));
ggml_tensor * inp_pos = build_inp_pos();
llm_graph_input_attn_k_iswa * inp_attn = build_attn_inp_k_iswa();
// no LeRoPE in the draft
const auto rope = [&](ggml_tensor * t) {
return ggml_rope_ext(ctx0, t, inp_pos, nullptr, n_rot, rope_type, n_ctx_orig, freq_base, freq_scale,
ext_factor, attn_factor, beta_fast, beta_slow);
};
const auto build_kv = [&](ggml_tensor * x, int il) -> std::pair<ggml_tensor *, ggml_tensor *> {
const auto & layer = model.layers[il];
ggml_tensor * kv_cmpr_pe = ggml_mul_mat(ctx0, layer.wkv_a_mqa, x);
cb(kv_cmpr_pe, "kv_cmpr_pe", il);
ggml_tensor * kv_cmpr = ggml_view_2d(ctx0, kv_cmpr_pe, kv_lora_rank, n_tokens,
ggml_row_size(kv_cmpr_pe->type, kv_lora_rank + n_rot), 0);
kv_cmpr = build_norm(kv_cmpr, layer.attn_kv_a_norm, nullptr, LLM_NORM_RMS, il);
kv_cmpr = ggml_reshape_3d(ctx0, kv_cmpr, kv_lora_rank, 1, n_tokens);
cb(kv_cmpr, "kv_cmpr", il);
ggml_tensor * k_pe = ggml_view_3d(ctx0, kv_cmpr_pe, n_rot, 1, n_tokens,
ggml_row_size(kv_cmpr_pe->type, kv_lora_rank + n_rot),
ggml_row_size(kv_cmpr_pe->type, kv_lora_rank + n_rot),
ggml_row_size(kv_cmpr_pe->type, kv_lora_rank));
k_pe = rope(k_pe);
cb(k_pe, "k_pe", il);
return { ggml_concat(ctx0, kv_cmpr, k_pe, 0), kv_cmpr };
};
ASSERT_EMBD_OR_TOKEN(ubatch);
if (ubatch.embd) {
auto inp = std::make_unique<llm_graph_input_embd>(n_embd_inp);
inp->embd = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd_inp, n_tokens);
ggml_set_input(inp->embd);
ggml_tensor * inp_target = inp->embd;
cb(inp_target, "inp_target_features", -1);
res->add_input(std::move(inp));
ggml_tensor * inp_g = build_lora_mm(model.fc, inp_target, model.fc_s);
inp_g = build_norm(inp_g, model.output_norm_enc, nullptr, LLM_NORM_RMS, -1);
cb(inp_g, "main_x", -1);
for (int il = 0; il < n_layer; ++il) {
ggml_tensor * kv = build_kv(inp_g, il).first;
cb(kv, "kv_injected", il);
if (inp_attn->self_k_rot_swa) {
kv = llama_mul_mat_hadamard(ctx0, kv, inp_attn->self_k_rot_swa);
}
ggml_build_forward_expand(gf, inp_attn->mctx->get_swa()->cpy_k(ctx0, kv, inp_attn->get_k_idxs_swa(), il));
}
res->t_embd = inp_g;
ggml_build_forward_expand(gf, inp_g);
return;
}
auto * tok_embd = model.tok_embd;
if (tok_embd == nullptr) {
GGML_ASSERT(cparams.ctx_other != nullptr);
const auto * model_other = llama_get_model(cparams.ctx_other);
GGML_ASSERT(model_other->tok_embd != nullptr && "DSpark decoder requires the target model's token embeddings");
tok_embd = model_other->tok_embd;
}
auto inp = std::make_unique<llm_graph_input_embd>(n_embd);
inp->tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens);
ggml_set_input(inp->tokens);
ggml_tensor * inp_tokens = inp->tokens;
res->add_input(std::move(inp));
const int64_t n_blocks = ubatch.n_seqs_unq;
const int64_t block_drafts = n_blocks > 0 ? n_tokens / n_blocks : 0;
ggml_tensor * inpL;
if (block_drafts > 0 && n_tokens % n_blocks == 0 && block_drafts <= (int64_t) hparams.dflash_block_size) {
// slot k > 0 uses mask row k - 1
ggml_tensor * anchors = ggml_view_2d(ctx0, inp_tokens, 1, n_blocks, (size_t) block_drafts * inp_tokens->nb[0], 0);
anchors = ggml_get_rows(ctx0, tok_embd, ggml_cont_1d(ctx0, anchors, n_blocks));
inpL = ggml_reshape_3d(ctx0, anchors, n_embd, 1, n_blocks);
if (block_drafts > 1) {
ggml_tensor * masks = ggml_view_3d(ctx0, model.dspark_mask_embd, n_embd, block_drafts - 1, 1,
model.dspark_mask_embd->nb[1], model.dspark_mask_embd->nb[2], 0);
masks = ggml_repeat_4d(ctx0, ggml_cast(ctx0, masks, GGML_TYPE_F32), n_embd, block_drafts - 1, n_blocks, 1);
inpL = ggml_concat(ctx0, inpL, masks, 1);
}
inpL = ggml_reshape_2d(ctx0, inpL, n_embd, n_tokens);
} else {
// not a draft block (e.g. graph reserve)
inpL = ggml_get_rows(ctx0, tok_embd, inp_tokens);
}
cb(inpL, "inp_noise_embd", -1);
build_dspark_block_mask(*this, inp_attn);
for (int il = 0; il < n_layer; ++il) {
const auto & layer = model.layers[il];
ggml_tensor * cur = build_norm(inpL, layer.attn_norm, nullptr, LLM_NORM_RMS, il);
cb(cur, "attn_norm", il);
{
ggml_tensor * q = ggml_mul_mat(ctx0, layer.wq, cur);
cb(q, "q", il);
// {n_embd_head_nope, n_tokens, n_head}
ggml_tensor * q_nope = ggml_view_3d(ctx0, q, n_embd_head_nope, n_head, n_tokens,
ggml_row_size(q->type, n_embd_head_k_mla),
ggml_row_size(q->type, n_embd_head_k_mla) * n_head, 0);
q_nope = ggml_permute(ctx0, q_nope, 0, 2, 1, 3);
// {kv_lora_rank, n_head, n_tokens}
ggml_tensor * q_nope_absorbed = ggml_mul_mat(ctx0, layer.wk_b, q_nope);
q_nope_absorbed = ggml_permute(ctx0, q_nope_absorbed, 0, 2, 1, 3);
cb(q_nope_absorbed, "q_nope_absorbed", il);
ggml_tensor * q_pe = ggml_view_3d(ctx0, q, n_rot, n_head, n_tokens,
ggml_row_size(q->type, n_embd_head_k_mla),
ggml_row_size(q->type, n_embd_head_k_mla) * n_head,
ggml_row_size(q->type, n_embd_head_nope));
q_pe = rope(q_pe);
cb(q_pe, "q_pe", il);
ggml_tensor * Qcur = ggml_concat(ctx0, q_nope_absorbed, q_pe, 0);
cb(Qcur, "Qcur", il);
auto [Kcur, Vcur] = build_kv(cur, il);
cb(Kcur, "Kcur", il);
// apply wv_b after attn, so the K rotation is already undone
ggml_tensor * out = build_attn(inp_attn,
nullptr, nullptr, nullptr,
Qcur, Kcur, Vcur,
nullptr, nullptr, nullptr,
kq_scale, il);
cb(out, "attn_latent", il);
// {kv_lora_rank, n_tokens, n_head} -> {n_embd_head_v_mla, n_tokens, n_head}
out = ggml_reshape_3d(ctx0, out, kv_lora_rank, n_head, n_tokens);
out = ggml_mul_mat(ctx0, layer.wv_b, ggml_permute(ctx0, out, 0, 2, 1, 3));
out = ggml_cont_2d(ctx0, ggml_permute(ctx0, out, 0, 2, 1, 3), n_embd_head_v_mla*n_head, n_tokens);
cur = build_lora_mm(layer.wo, out, layer.wo_s);
cb(cur, "attn_out", il);
}
ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpL);
cb(ffn_inp, "ffn_inp", il);
cur = build_norm(ffn_inp, layer.ffn_norm, nullptr, LLM_NORM_RMS, il);
cb(cur, "ffn_norm", il);
cur = build_ffn(cur,
layer.ffn_up, nullptr, nullptr,
layer.ffn_gate, nullptr, nullptr,
layer.ffn_down, nullptr, nullptr,
nullptr, LLM_FFN_SILU, LLM_FFN_PAR, il);
cb(cur, "ffn_out", il);
inpL = ggml_add(ctx0, cur, ffn_inp);
cb(inpL, "l_out", il);
}
ggml_tensor * cur = build_norm(inpL, model.output_norm, nullptr, LLM_NORM_RMS, -1);
cb(cur, "result_norm", -1);
res->t_embd = cur;
auto * output = model.output;
auto * output_s = model.output_s;
if (output == nullptr) {
GGML_ASSERT(cparams.ctx_other != nullptr);
const auto * model_other = llama_get_model(cparams.ctx_other);
GGML_ASSERT(model_other->output != nullptr && "DSpark decoder requires the target model's output projection");
output = model_other->output;
output_s = model_other->output_s;
}
ggml_tensor * logits = build_lora_mm(output, cur, output_s);
// softcap does not apply to the markov bias
if (hparams.f_final_logit_softcapping) {
logits = ggml_scale(ctx0, logits, 1.0f / hparams.f_final_logit_softcapping);
logits = ggml_tanh(ctx0, logits);
logits = ggml_scale(ctx0, logits, hparams.f_final_logit_softcapping);
}
cb(logits, "result_output", -1);
res->t_logits = logits;
if (model.dspark_markov_w1) {
build_dspark_markov_head(*this, model, inp_tokens, cur);
}
}
std::unique_ptr<llm_graph_context> llama_model_dflash::build_arch_graph(const llm_graph_params & params) const {
switch (params.gtype) {
case LLM_GRAPH_TYPE_ENCODER:
return std::make_unique<graph<true>>(*this, params);
case LLM_GRAPH_TYPE_DEFAULT:
case LLM_GRAPH_TYPE_DECODER:
if (hparams.n_lora_kv > 0) {
return std::make_unique<graph_mistral_large4>(*this, params);
}
if (hparams.dsv4_hc_mult > 0 && hparams.dsv4_hc_single_pass) {
return std::make_unique<graph_dsv41>(*this, params);
}