tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-17 21:51:27 -04:00
Code Issues Packages Projects Releases Wiki Activity

llama : more consistent names of count variables (#5994)

Browse Source 
* llama : more consistent names of count variables

ggml-ci

* llama : n_parallel -> n_seq_max

* common : fix param name

* examples : fix param name
This commit is contained in:
Georgi Gerganov
2024-03-11 17:49:47 +02:00
committed by GitHub
parent 83796e62bc
commit 05b06210c9
8 changed files with 35 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/batched-bench/batched-bench.cpp
View File

@@ -106,7 +106,7 @@ int main(int argc, char ** argv) {
ctx_params.n_threads_batch = params.n_threads_batch == -1 ? params.n_threads : params.n_threads_batch;
// ensure enough sequences are available
ctx_params.n_parallel = *std::max_element(n_pl.begin(), n_pl.end());
ctx_params.n_seq_max = *std::max_element(n_pl.begin(), n_pl.end());
llama_context * ctx = llama_new_context_with_model(model, ctx_params);

2
examples/batched/batched.cpp
View File

@@ -80,7 +80,7 @@ int main(int argc, char ** argv) {
ctx_params.seed = 1234;
ctx_params.n_ctx = n_kv_req;
ctx_params.n_batch = std::max(n_len, n_parallel);
ctx_params.n_parallel = n_parallel;
ctx_params.n_seq_max = n_parallel;
ctx_params.n_threads = params.n_threads;
ctx_params.n_threads_batch = params.n_threads_batch == -1 ? params.n_threads : params.n_threads_batch;

1
examples/main/main.cpp
View File

@@ -878,6 +878,7 @@ int main(int argc, char ** argv) {
const auto line_pfx = ::llama_tokenize(ctx, params.input_prefix, false, true);
const auto line_inp = ::llama_tokenize(ctx, buffer, false, false);
const auto line_sfx = ::llama_tokenize(ctx, params.input_suffix, false, true);
LOG("input tokens: %s\n", LOG_TOKENS_TOSTR_PRETTY(ctx, line_inp).c_str());
embd_inp.insert(embd_inp.end(), line_pfx.begin(), line_pfx.end());

6
examples/perplexity/perplexity.cpp
View File

@@ -841,7 +841,7 @@ static void hellaswag_score(llama_context * ctx, const gpt_params & params) {
const int n_batch = params.n_batch;
const int max_tasks_per_batch = 32;
const int max_seq = std::min(4*max_tasks_per_batch, (int) llama_n_max_seq(ctx));
const int max_seq = std::min(4*max_tasks_per_batch, (int) llama_n_seq_max(ctx));
llama_batch batch = llama_batch_init(n_ctx, 0, max_seq);
@@ -1118,7 +1118,7 @@ static void winogrande_score(llama_context * ctx, const gpt_params & params) {
const int n_batch = params.n_batch;
const int max_tasks_per_batch = 128;
const int max_seq = std::min(2*max_tasks_per_batch, (int) llama_n_max_seq(ctx));
const int max_seq = std::min(2*max_tasks_per_batch, (int) llama_n_seq_max(ctx));
llama_batch batch = llama_batch_init(n_ctx, 0, max_seq);
@@ -1470,7 +1470,7 @@ static void multiple_choice_score(llama_context * ctx, const gpt_params & params
const int n_batch = params.n_batch;
const int max_tasks_per_batch = 32;
const int max_seq = std::min(4*max_tasks_per_batch, (int) llama_n_max_seq(ctx));
const int max_seq = std::min(4*max_tasks_per_batch, (int) llama_n_seq_max(ctx));
llama_batch batch = llama_batch_init(n_ctx, 0, max_seq);