root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-27 03:44:35 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

fix vram calculation for starcoder

Browse Source
This commit is contained in:
Meng Zhang 2023-09-15 13:52:43 +08:00
parent dac31da489
commit 4420cff654
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
llama.cpp
View File

@ -2277,8 +2277,11 @@ static void llm_load_tensors(
if (backend == GGML_BACKEND_GPU) {
vram_weights +=
ggml_nbytes(layer.attn_norm) + ggml_nbytes(layer.attn_norm_b) +
ggml_nbytes(layer.wqkv) + ggml_nbytes(layer.wo) +
ggml_nbytes(layer.w2) + ggml_nbytes(layer.w3);
ggml_nbytes(layer.wqkv) + ggml_nbytes(layer.bqkv) +
ggml_nbytes(layer.wo) + ggml_nbytes(layer.bo) +
ggml_nbytes(layer.ffn_norm) + ggml_nbytes(layer.ffn_norm_b) +
ggml_nbytes(layer.w2) + ggml_nbytes(layer.b2) +
ggml_nbytes(layer.w3) + ggml_nbytes(layer.b3);
}
}
} break;