vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

debugging

Browse source
This commit is contained in:
mike dupont 2023-11-22 17:43:16 -05:00
parent b598cf84fa
commit c683f2c76a
3 changed files with 24 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

5
Makefile
View file

@ -582,7 +582,7 @@ clean:
# Examples
#
main: examples/main/main.cpp ggml.o llama.o $(COMMON_DEPS) console.o grammar-parser.o $(OBJS)
main: examples/main/main.cpp ggml.o llama.o $(COMMON_DEPS) console.o grammar-parser.o $(OBJS)
$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
@echo
@echo '==== Run ./main -h for help. ===='
@ -678,6 +678,9 @@ common/build-info.cpp: $(wildcard .git/index) scripts/build-info.sh
build-info.o: common/build-info.cpp
$(CXX) $(CXXFLAGS) -c $(filter-out %.h,$^) -o $@
#print.o: print.cpp # print.hpp
# $(CXX) $(CXXFLAGS) -c $(filter-out %.h,$^) -o $@
#
# Tests
#

19
examples/main/main.cpp
View file

@ -31,6 +31,8 @@
#pragma warning(disable: 4244 4267) // possible loss of data
#endif
#include "print.hpp"
static llama_context ** g_ctx;
static llama_model ** g_model;
static gpt_params * g_params;
@ -105,7 +107,9 @@ int main(int argc, char ** argv) {
gpt_params params;
g_params = &params;
using Td = type_descriptor<gpt_params>;
//using Td = type_descriptor<gpt_params>;
print_fields(g_params);
//constexpr auto tbl = descriptor::get_attribute<gpt_params>(Td{});
//constexpr auto tbl_name = REFL_MAKE_CONST_STRING(tbl.name);
@ -180,6 +184,9 @@ int main(int argc, char ** argv) {
g_model = &model;
g_ctx = &ctx;
print_fields(g_model);
print_fields(g_ctx);
// load the model and apply lora adapter, if any
LOG("%s: load the model and apply lora adapter, if any\n", __func__);
std::tie(model, ctx) = llama_init_from_gpt_params(params);
@ -239,6 +246,8 @@ int main(int argc, char ** argv) {
std::vector<llama_token> embd_inp;
if (params.interactive_first || params.instruct || params.chatml || !params.prompt.empty() || session_tokens.empty()) {
LOG("tokenize the prompt\n");
if (params.chatml) {
@ -258,7 +267,7 @@ int main(int argc, char ** argv) {
embd_inp.push_back(llama_token_bos(model));
LOG("embd_inp was considered empty and bos was added: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx, embd_inp).c_str());
}
//print_fields(embd_inp);
// Tokenize negative prompt
std::vector<llama_token> guidance_inp;
int guidance_offset = 0;
@ -283,6 +292,7 @@ int main(int argc, char ** argv) {
return 1;
}
//print_fields(session_tokens);
// debug message about similarity of saved session, if applicable
size_t n_matching_session_tokens = 0;
if (!session_tokens.empty()) {
@ -478,7 +488,8 @@ int main(int argc, char ** argv) {
std::vector<llama_token> embd_guidance;
struct llama_sampling_context * ctx_sampling = llama_sampling_init(sparams);
print_fields(ctx_sampling);
while ((n_remain != 0 && !is_antiprompt) || params.interactive) {
// predict
if (!embd.empty()) {
@ -487,6 +498,7 @@ int main(int argc, char ** argv) {
int max_embd_size = n_ctx - 4;
// Ensure the input doesn't exceed the context size by truncating embd if necessary.
//print_fields(embd);
if ((int) embd.size() > max_embd_size) {
const int skipped_tokens = (int) embd.size() - max_embd_size;
embd.resize(max_embd_size);
@ -513,6 +525,7 @@ int main(int argc, char ** argv) {
LOG("context full, swapping: n_past = %d, n_left = %d, n_ctx = %d, n_keep = %d, n_discard = %d\n",
n_past, n_left, n_ctx, params.n_keep, n_discard);
print_fields(ctx);
llama_kv_cache_seq_rm (ctx, 0, params.n_keep + 1 , params.n_keep + n_discard + 1);
llama_kv_cache_seq_shift(ctx, 0, params.n_keep + 1 + n_discard, n_past, -n_discard);

11
llama.h
View file

@ -115,13 +115,8 @@ extern "C" {
};
typedef struct llama_token_data : refl::attr::usage::type{
llama_token_data( llama_token id,
float logit,
float p):
id( id),
logit(logit),
p(p){
}
llama_token_data( llama_token id, float logit, float p):
id( id),logit(logit),p(p){ }
llama_token id; // token id
float logit; // log-odds of the token
float p; // probability of the token
@ -833,4 +828,6 @@ const std::vector<std::pair<std::string, struct ggml_tensor *>> & llama_internal
#endif // LLAMA_API_INTERNAL
template<typename T> void print_fields(const T& obj);
#endif // LLAMA_H