CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-04 09:32:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

imatrix : fix segfault when using a single chunk per batch

Browse Source
This commit is contained in:
Francis Couture-Harpin
2024-08-20 15:35:56 -04:00
parent bce54642c8
commit 347247a24e
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/imatrix/imatrix.cpp
View File

@@ -564,7 +564,7 @@ static bool compute_imatrix(llama_context * ctx, const gpt_params & params, cons
if (params.compute_ppl) { if (params.compute_ppl) {
const int first = n_ctx/2; const int first = n_ctx/2;
for (int seq = 0; seq < n_seq_batch; seq++) { for (int seq = 0; seq < n_seq_batch; seq++) {
const float * all_logits = num_batches > 1 ? logits.data() : llama_get_logits_ith(ctx, seq*n_ctx + first); const float * all_logits = num_batches > 1 ? logits.data() : llama_get_logits_ith(ctx, seq*n_ctx);
llama_token * tokens_data = tokens.data() + start + seq*n_ctx + first; llama_token * tokens_data = tokens.data() + start + seq*n_ctx + first;