happyz
/
gemma.cpp
mirror of https://github.com/google/gemma.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Internal change. 

PiperOrigin-RevId: 765218260
This commit is contained in:
The gemma.cpp Authors 2025-05-30 09:18:08 -07:00 committed by Copybara-Service
parent 1e8642f8f4
commit 9c3e089b09
2 changed files with 44 additions and 46 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
gemma/configs.cc
View File

@ -572,7 +572,8 @@ ModelConfig::ModelConfig(const Model model, Type weight,
static Model FindModel(const std::string& specifier) { static Model FindModel(const std::string& specifier) {
Model found_model = Model::UNKNOWN; Model found_model = Model::UNKNOWN;
ForEachModel([&](Model model) { ForEachModel([&](Model model) {
const char* prefix = ModelPrefix(model); // Some model names are prefixes of other model names
const std::string prefix = std::string(ModelPrefix(model)) + "-";
if (specifier.rfind(prefix, 0) == 0) { // Starts with prefix. if (specifier.rfind(prefix, 0) == 0) { // Starts with prefix.
// We only expect one match. // We only expect one match.
HWY_ASSERT_M(found_model == Model::UNKNOWN, specifier.c_str()); HWY_ASSERT_M(found_model == Model::UNKNOWN, specifier.c_str());

87
gemma/gemma-inl.h
View File

@ -176,7 +176,7 @@ HWY_NOINLINE void GriffinRecurrent(const QueriesPos& queries_pos,
Sigmoid(gate_x + head_offset, kHeadDim); Sigmoid(gate_x + head_offset, kHeadDim);
Sigmoid(a + head_offset, kHeadDim); Sigmoid(a + head_offset, kHeadDim);
const auto fn_mul = [](D d, hn::Vec<D> x, hn::Vec<D> gate_x) const auto fn_mul = [](D d, hn::Vec<D> x, hn::Vec<D> gate_x)
HWY_ATTR { return hn::Mul(x, gate_x); }; HWY_ATTR { return hn::Mul(x, gate_x); };
hn::Transform1(D(), a + head_offset, kHeadDim, hn::Transform1(D(), a + head_offset, kHeadDim,
layer_weights->griffin.a.PackedScale1() + head_offset, layer_weights->griffin.a.PackedScale1() + head_offset,
fn_mul); fn_mul);
@ -424,51 +424,49 @@ class GemmaAttention {
const size_t kHeadGroups = layer_config_.heads / layer_config_.kv_heads; const size_t kHeadGroups = layer_config_.heads / layer_config_.kv_heads;
// For each head (token, query), compute Q.K, softmax, and weighted V. // For each head (token, query), compute Q.K, softmax, and weighted V.
pool_.Run(0, layer_config_.heads * num_interleaved, pool_.Run(
[&](uint64_t task, size_t /*thread*/) HWY_ATTR { 0, layer_config_.heads * num_interleaved,
const size_t head = task % layer_config_.heads; [&](uint64_t task, size_t /*thread*/) HWY_ATTR {
const size_t interleaved_idx = task / layer_config_.heads; const size_t head = task % layer_config_.heads;
const size_t query_idx = interleaved_idx % num_queries_; const size_t interleaved_idx = task / layer_config_.heads;
const size_t batch_idx = interleaved_idx / num_queries_; const size_t query_idx = interleaved_idx % num_queries_;
const size_t qkv_dim = layer_config_.qkv_dim; const size_t batch_idx = interleaved_idx / num_queries_;
const size_t head_offset = (head / kHeadGroups) * qkv_dim * 2; const size_t qkv_dim = layer_config_.qkv_dim;
const size_t head_offset = (head / kHeadGroups) * qkv_dim * 2;
float* HWY_RESTRICT q = float* HWY_RESTRICT q =
activations_.q.Row(interleaved_idx) + head * q_stride_; activations_.q.Row(interleaved_idx) + head * q_stride_;
float* HWY_RESTRICT att = float* HWY_RESTRICT att = activations_.att.Row(interleaved_idx) +
activations_.att.Row(interleaved_idx) + head * activations_.seq_len;
head * activations_.seq_len; float* HWY_RESTRICT att_out =
float* HWY_RESTRICT att_out = activations_.att_out.Row(interleaved_idx) + head * qkv_dim;
activations_.att_out.Row(interleaved_idx) + head * qkv_dim;
// Make strided views into the kv cache entries for the current // Make strided views into the kv cache entries for the current
// query and head. // query and head.
KVCache& kv_cache = kv_caches_[query_idx]; KVCache& kv_cache = kv_caches_[query_idx];
const size_t kv_head_offset = const size_t kv_head_offset =
layer_ * cache_layer_size_ + head_offset; layer_ * cache_layer_size_ + head_offset;
MatPtrT<float> k("k_view", MatPtrT<float> k("k_view", Extents2D(kv_cache.seq_len, qkv_dim));
Extents2D(kv_cache.seq_len, qkv_dim)); k.SetPtr(kv_cache.kv_cache.get() + kv_head_offset,
k.SetPtr(kv_cache.kv_cache.get() + kv_head_offset, /*stride=*/cache_pos_size_);
/*stride=*/cache_pos_size_); MatPtrT<float> v("v_view", Extents2D(kv_cache.seq_len, qkv_dim));
MatPtrT<float> v("v_view", v.SetPtr(kv_cache.kv_cache.get() + kv_head_offset + qkv_dim,
Extents2D(kv_cache.seq_len, qkv_dim)); /*stride=*/cache_pos_size_);
v.SetPtr(kv_cache.kv_cache.get() + kv_head_offset + qkv_dim,
/*stride=*/cache_pos_size_);
// Find the token position in the query and calculate the range // Find the token position in the query and calculate the range
// of cache positions to attend to. // of cache positions to attend to.
const size_t pos = queries_pos_[query_idx] + batch_idx; const size_t pos = queries_pos_[query_idx] + batch_idx;
const size_t start_pos = StartPos(pos, layer_); const size_t start_pos = StartPos(pos, layer_);
size_t last_pos = pos; size_t last_pos = pos;
const size_t prefix_end = queries_prefix_end_[query_idx]; const size_t prefix_end = queries_prefix_end_[query_idx];
if (prefix_end > 0 && prefix_end - 1 > last_pos) { if (prefix_end > 0 && prefix_end - 1 > last_pos) {
// last_pos in QDotK and WeightedSumV is inclusive. // last_pos in QDotK and WeightedSumV is inclusive.
last_pos = prefix_end - 1; last_pos = prefix_end - 1;
} }
SingleDotSoftmaxWeightedSum(q, k, v, att, att_out, query_scale, SingleDotSoftmaxWeightedSum(q, k, v, att, att_out, query_scale, pos,
pos, start_pos, last_pos); start_pos, last_pos);
}); });
} }
private: private:
@ -1510,8 +1508,7 @@ void GenerateSingleT(const ModelStore& model,
} }
template <typename T> template <typename T>
void GenerateBatchT(const ModelStore& model, void GenerateBatchT(const ModelStore& model, const ModelWeightsPtrs<T>& weights,
const ModelWeightsPtrs<T>& weights,
const RuntimeConfig& runtime_config, const RuntimeConfig& runtime_config,
const QueriesPromptTokens& queries_prompt, const QueriesPromptTokens& queries_prompt,
const QueriesPos& queries_pos, const QueriesPos& queries_pos,
@ -1536,7 +1533,7 @@ void GenerateBatchT(const ModelStore& model,
qbatch_size); qbatch_size);
QueriesPos qbatch_pos(&queries_pos[qbatch_start], qbatch_size); QueriesPos qbatch_pos(&queries_pos[qbatch_start], qbatch_size);
const QueriesPos qbatch_prefix_end(&queries_prefix_end[qbatch_start], const QueriesPos qbatch_prefix_end(&queries_prefix_end[qbatch_start],
qbatch_size); qbatch_size);
const KVCaches qbatch_kv(&kv_caches[qbatch_start], qbatch_size); const KVCaches qbatch_kv(&kv_caches[qbatch_start], qbatch_size);
GenerateT<T>(model, weights, activations, runtime_config, qbatch_prompts, GenerateT<T>(model, weights, activations, runtime_config, qbatch_prompts,
qbatch_pos, qbatch_prefix_end, qbatch_start, qbatch_kv, qbatch_pos, qbatch_prefix_end, qbatch_start, qbatch_kv,