vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-15 03:19:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'master' into concedo

Browse source 
# Conflicts:
#	Makefile
This commit is contained in:
Concedo 2023-04-08 17:42:09 +08:00
commit 0b904e12db
10 changed files with 430 additions and 33 deletions
Download patch file Download diff file Expand all files Collapse all files

22
llama.cpp
View file

@ -1238,19 +1238,13 @@ static llama_vocab::id llama_sample_top_p_top_k(
}
}
if (top_k > 0 && top_k < n_logits) {
sample_top_k(logits_id, top_k);
}
float maxl = -std::numeric_limits<float>::infinity();
for (const auto & kv : logits_id) {
maxl = Max(maxl, kv.first);
}
sample_top_k(logits_id, top_k > 0 ? Min(top_k, n_logits) : n_logits);
// compute probs for the top k tokens
std::vector<float> probs;
probs.reserve(logits_id.size());
float maxl = logits_id[0].first;
double sum = 0.0;
for (const auto & kv : logits_id) {
const float p = expf(kv.first - maxl);
@ -1273,16 +1267,11 @@ static llama_vocab::id llama_sample_top_p_top_k(
break;
}
}
cumsum = 1.0/cumsum;
for (int i = 0; i < (int) probs.size(); i++) {
probs[i] *= cumsum;
}
}
//printf("\n");
//for (int i = 0; i < (int) 10; i++) {
// printf("%d: '%s' %f\n", i, vocab.id_to_token.at(logits_id[i].second).c_str(), probs[i]);
// printf("%d: '%s' %f\n", i, lctx.vocab.id_to_token.at(logits_id[i].second).tok.c_str(), probs[i]);
//}
//printf("\n\n");
//exit(0);
@ -1865,3 +1854,8 @@ const char * llama_print_system_info(void) {
return s.c_str();
}
// For internal test use
std::unordered_map<std::string, struct ggml_tensor *>& llama_internal_get_tensor_map(struct llama_context * ctx) {
return ctx->model.tensors;
}