tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-30 04:45:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

wip

Browse Source
This commit is contained in:
Georgi Gerganov
2024-11-16 10:04:49 +02:00
parent f245cc28d4
commit f7b0233eca
1 changed files with 8 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
examples/llama-bench/llama-bench.cpp
View File

@ -1337,7 +1337,14 @@ static void test_prompt(llama_context * ctx, int n_prompt, int n_batch, int n_th
for (int i = 1; i < n_tokens; i++) { for (int i = 1; i < n_tokens; i++) {
tokens[i] = std::rand() % n_vocab; tokens[i] = std::rand() % n_vocab;
} }
llama_decode(ctx, llama_batch_get_one(tokens.data(), n_tokens)); auto batch = llama_batch_get_one(tokens.data(), n_tokens);
int8_t logits[512];
for (int i = 0; i < n_tokens; i++) {
logits[i] = 1;
}
batch.logits = logits;
llama_decode(ctx, batch);
n_processed += n_tokens; n_processed += n_tokens;
} }