vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-11 01:24:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	.github/workflows/build.yml
#	.github/workflows/docker.yml
#	Makefile
#	README-sycl.md
#	README.md
#	ci/run.sh
#	ggml-cuda.cu
#	ggml.c
#	grammars/README.md
#	scripts/get-wikitext-2.sh
#	scripts/hf.sh
#	scripts/sync-ggml.last
#	tests/test-backend-ops.cpp
#	tests/test-grammar-integration.cpp
#	tests/test-json-schema-to-grammar.cpp
This commit is contained in:
Concedo 2024-04-14 21:18:39 +08:00
commit 9a25d77cc1
58 changed files with 6529 additions and 6121 deletions
Download patch file Download diff file Expand all files Collapse all files

20
common/common.cpp
View file

@ -1746,6 +1746,8 @@ struct llama_context_params llama_context_params_from_gpt_params(const gpt_param
cparams.yarn_orig_ctx = params.yarn_orig_ctx;
cparams.pooling_type = params.pooling_type;
cparams.defrag_thold = params.defrag_thold;
cparams.cb_eval = params.cb_eval;
cparams.cb_eval_user_data = params.cb_eval_user_data;
cparams.offload_kqv = !params.no_kv_offload;
cparams.type_k = kv_cache_type_from_str(params.cache_type_k);
@ -2193,7 +2195,7 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
params.sparams.logit_bias[llama_token_eos(model)] = -INFINITY;
}
{
if (params.warmup) {
LOG("warming up the model with an empty run\n");
std::vector<llama_token> tmp = { llama_token_bos(model), llama_token_eos(model), };
@ -2213,23 +2215,23 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
std::vector<llama_token> llama_tokenize(
const struct llama_context * ctx,
const std::string & text,
bool add_bos,
bool special) {
return llama_tokenize(llama_get_model(ctx), text, add_bos, special);
bool add_special,
bool parse_special) {
return llama_tokenize(llama_get_model(ctx), text, add_special, parse_special);
}
std::vector<llama_token> llama_tokenize(
const struct llama_model * model,
const std::string & text,
bool add_bos,
bool special) {
bool add_special,
bool parse_special) {
// upper limit for the number of tokens
int n_tokens = text.length() + add_bos;
int n_tokens = text.length() + 2 * add_special;
std::vector<llama_token> result(n_tokens);
n_tokens = llama_tokenize(model, text.data(), text.length(), result.data(), result.size(), add_bos, special);
n_tokens = llama_tokenize(model, text.data(), text.length(), result.data(), result.size(), add_special, parse_special);
if (n_tokens < 0) {
result.resize(-n_tokens);
int check = llama_tokenize(model, text.data(), text.length(), result.data(), result.size(), add_bos, special);
int check = llama_tokenize(model, text.data(), text.length(), result.data(), result.size(), add_special, parse_special);
GGML_ASSERT(check == -n_tokens);
} else {
result.resize(n_tokens);