ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

llama.cpp : split llama_context_params into model and context params (#3301)

Browse source 
* llama.cpp : split llama_context_params into model and context params

ggml-ci

* fix metal build

* fix freq_base/scale default to model value

* llama-bench : keep the same model between tests when possible

* move n_threads to llama_context_params, add n_threads_batch

* fix mpi build

* remove kv_size(), cuda scratch fixes

* remove low-vram option

* add n_threads_batch to system info, refactor to get_system_info()

* add documentation about --threads-batch to the READMEs

* llama-bench fix

* main : fix rope freq/scale warning

* llama.cpp : add llama_get_model
common : add llama_tokenize from model

* remove duplicated ctx/model functions

ggml-ci

* cuda : print total VRAM used
This commit is contained in:
slaren 2023-09-28 21:42:38 +02:00 • committed by GitHub
parent 0512d66670
commit 16bc66d947
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
27 changed files with 713 additions and 633 deletions
Download patch file Download diff file Expand all files Collapse all files

10
common/train.cpp
View file

@ -858,7 +858,7 @@ size_t tokenize_file(
out_tokens.resize(buf.size() + n_max_tokens_overhead);
int n_tokens = llama_tokenize(
lctx,
llama_get_model(lctx),
buf.data(),
(int) buf.size(),
out_tokens.data(),
@ -867,7 +867,7 @@ size_t tokenize_file(
if (n_tokens < 0) {
out_tokens.resize(-n_tokens);
n_tokens = llama_tokenize(
lctx,
llama_get_model(lctx),
buf.data(),
(int) buf.size(),
out_tokens.data(),
@ -920,7 +920,7 @@ size_t tokenize_file(
size_t found_max_sample_size = 0;
size_t max_token_text_size = 0;
int n_vocab = llama_n_vocab(lctx);
int n_vocab = llama_n_vocab(llama_get_model(lctx));
for (llama_token token=0; token < n_vocab; ++token) {
max_token_text_size = std::max(
max_token_text_size,
@ -961,7 +961,7 @@ size_t tokenize_file(
// tokenize the sample
tok_sample.resize(buf_sample.size() + n_max_tokens_overhead);
int n_tokens = llama_tokenize(lctx,
int n_tokens = llama_tokenize(llama_get_model(lctx),
buf_sample.data(),
(int) buf_sample.size(),
tok_sample.data(),
@ -969,7 +969,7 @@ size_t tokenize_file(
false);
if (n_tokens < 0) {
tok_sample.resize(-n_tokens);
n_tokens = llama_tokenize(lctx,
n_tokens = llama_tokenize(llama_get_model(lctx),
buf_sample.data(),
(int) buf_sample.size(),
tok_sample.data(),