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

Add super wip scripts for multimodal granite gguf

Browse source 
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
This commit is contained in:
Alex-Brooks 2025-01-16 14:54:31 -07:00
parent d774ab3acc
commit 6ccf234031
4 changed files with 119 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ggml/src/ggml-cpu/ggml-cpu.c
View file

@ -8515,7 +8515,9 @@ static void ggml_compute_forward_get_rows_f32(
const int64_t i10 = (i - i12*ne11*ne10 - i11*ne10);
const int64_t i01 = *(int32_t *) ((char *) src1->data + i10*nb10 + i11*nb11 + i12*nb12);
GGML_ASSERT(i01 >= 0 && i01 < ne01);
// Copying this out for a bit while investigating due to issues like:
// https://github.com/ggerganov/llama.cpp/issues/10157
// GGML_ASSERT(i01 >= 0 && i01 < ne01);
ggml_vec_cpy_f32(nc,
(float *) ((char *) dst->data + i10*nb1 + i11*nb2 + i12*nb3),