8000 llama : add support for jina-reranker-v2 by CISC · Pull Request #13900 · ggml-org/llama.cpp · GitHub
[go: up one dir, main page]
Skip to content

llama : add support for jina-reranker-v2 #13900

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
May 29, 2025
Merged

llama : add support for jina-reranker-v2 #13900

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
725108b
add jina qkv tensors
CISC May 29, 2025
928fea0
support conversion without sentencepiece.bpe.model
CISC May 29, 2025
024b68d
allow merged qkv tensor in place of split ones
CISC May 29, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
allow merged qkv tensor in place of split ones
  • Loading branch information
@CISC
CISC authored May 29, 2025
commit 024b68ddd3ad24f52e4d3740d56a498e3d1f6377
1 change: 1 addition & 0 deletions src/llama-arch.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -450,6 +450,7 @@ static const std::map<llm_arch, std::map<llm_tensor, const char *>> LLM_TENSOR_N
{ LLM_TENSOR_TOKEN_TYPES, "token_types" },
{ LLM_TENSOR_POS_EMBD, "position_embd" },
{ LLM_TENSOR_ATTN_OUT_NORM, "blk.%d.attn_output_norm" },
{ LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
{ LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
{ LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
{ LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
Expand Down
67 changes: 30 additions & 37 deletions src/llama-model.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2132,7 +2132,10 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
for (int i = 0; i < n_layer; ++i) {
auto & layer = layers[i];

if (arch == LLM_ARCH_BERT) {
layer.wqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, TENSOR_NOT_REQUIRED);
layer.bqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "bias", i), {n_embd + 2*n_embd_gqa}, TENSOR_NOT_REQUIRED);

if (!layer.wqkv) {
layer.wq = create_tensor(tn(LLM_TENSOR_ATTN_Q, "weight", i), {n_embd, n_embd}, 0);
layer.bq = create_tensor(tn(LLM_TENSOR_ATTN_Q, "bias", i), {n_embd}, 0);

Expand All @@ -2141,12 +2144,6 @@ bool llama_model::load_tensors(llama_model_loader & ml) {

layer.wv = create_tensor(tn(LLM_TENSOR_ATTN_V, "weight", i), {n_embd, n_embd_gqa}, 0);
layer.bv = create_tensor(tn(LLM_TENSOR_ATTN_V, "bias", i), {n_embd_gqa}, 0);
} else {
layer.wqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, 0);
}

if (arch == LLM_ARCH_NOMIC_BERT_MOE) {
layer.bqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "bias", i), {n_embd + 2*n_embd_gqa}, 0);
}

layer.wo = create_tensor(tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd}, 0);
Expand Down Expand Up @@ -5910,48 +5907,44 @@ struct llm_build_bert : public llm_graph_context {
ggml_tensor * Vcur;

// self-attention
if (model.arch == LLM_ARCH_BERT || model.arch == LLM_ARCH_JINA_BERT_V2) {
Qcur = ggml_add(ctx0, build_lora_mm(model.layers[il].wq, cur), model.layers[il].bq);

if (model.layers[il].attn_q_norm) {
Qcur = build_norm(Qcur,
model.layers[il].attn_q_norm,
model.layers[il].attn_q_norm_b,
LLM_NORM, il);
}

Kcur = ggml_add(ctx0, build_lora_mm(model.layers[il].wk, cur), model.layers[il].bk);

if (model.layers[il].attn_k_norm) {
Kcur = build_norm(Kcur,
model.layers[il].attn_k_norm,
model.layers[il].attn_k_norm_b,
LLM_NORM, il);
}

Vcur = ggml_add(ctx0, build_lora_mm(model.layers[il].wv, cur), model.layers[il].bv);

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);
Vcur = ggml_reshape_3d(ctx0, Vcur, n_embd_head, n_head_kv, n_tokens);
} else {
// compute Q and K and RoPE them
if (model.layers[il].wqkv) {
cur = build_lora_mm(model.layers[il].wqkv, cur);
cb(cur, "wqkv", il);

if (model.arch == LLM_ARCH_NOMIC_BERT_MOE) {
if (model.layers[il].bqkv) {
cur = ggml_add(ctx0, cur, model.layers[il].bqkv);
cb(cur, "bqkv", il);
}

Qcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd, n_tokens, cur->nb[1], 0*sizeof(float)*(n_embd)));
Kcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd)));
Vcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd + n_embd_gqa)));
} else {
Qcur = ggml_add(ctx0, build_lora_mm(model.layers[il].wq, cur), model.layers[il].bq);
Kcur = ggml_add(ctx0, build_lora_mm(model.layers[il].wk, cur), model.layers[il].bk);
Vcur = ggml_add(ctx0, build_lora_mm(model.layers[il].wv, cur), model.layers[il].bv);
}

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);
Vcur = ggml_reshape_3d(ctx0, Vcur, n_embd_head, n_head_kv, n_tokens);
if (model.layers[il].attn_q_norm) {
Qcur = build_norm(Qcur,
model.layers[il].attn_q_norm,
model.layers[il].attn_q_norm_b,
LLM_NORM, il);
}

if (model.layers[il].attn_k_norm) {
Kcur = build_norm(Kcur,
model.layers[il].attn_k_norm,
model.layers[il].attn_k_norm_b,
LLM_NORM, 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);
Vcur = ggml_reshape_3d(ctx0, Vcur, n_embd_head, n_head_kv, n_tokens);

// RoPE
if (model.arch == LLM_ARCH_NOMIC_BERT || model.arch == LLM_ARCH_NOMIC_BERT_MOE) {
Qcur = ggml_rope_ext(
ctx0, Qcur, inp_pos, nullptr,
n_rot, rope_type, n_ctx_orig, freq_base, freq_scale,
Expand Down
Loading
0