CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-01 09:01:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : avoid hardcoded special tokens

Browse Source
This commit is contained in:
Georgi Gerganov
2023-08-18 17:29:20 +03:00
parent 035d511457
commit 5d2656d670
11 changed files with 61 additions and 65 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/perplexity/perplexity.cpp
View File

@@ -63,7 +63,7 @@ void perplexity(llama_context * ctx, const gpt_params & params) {
// add BOS token for the first batch of each chunk
if (j == 0) {
tokens[batch_start] = llama_token_bos();
tokens[batch_start] = llama_token_bos(ctx);
}
if (llama_eval(ctx, tokens.data() + batch_start, batch_size, j * n_batch, params.n_threads)) {