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

update quantization types in switch-case of add_at and add1

Browse source
This commit is contained in:
xaedes 2023-05-01 14:30:29 +02:00
parent 72bcfb50c8
commit 1c4dc1e498
No known key found for this signature in database
GPG key ID: 30030EDD817EA2B1
1 changed files with 8 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

10
ggml.c
View file

@ -7974,7 +7974,10 @@ static void ggml_compute_forward_add1(
case GGML_TYPE_Q4_0:
case GGML_TYPE_Q4_1:
case GGML_TYPE_Q4_2:
case GGML_TYPE_Q4_3:
case GGML_TYPE_Q5_0:
case GGML_TYPE_Q5_1:
case GGML_TYPE_Q8_0:
case GGML_TYPE_Q8_1:
{
ggml_compute_forward_add1_q_f32(params, src0, src1, dst);
} break;
@ -8094,7 +8097,10 @@ static void ggml_compute_forward_add_at(
case GGML_TYPE_Q4_0:
case GGML_TYPE_Q4_1:
case GGML_TYPE_Q4_2:
case GGML_TYPE_Q4_3:
case GGML_TYPE_Q5_0:
case GGML_TYPE_Q5_1:
case GGML_TYPE_Q8_0:
case GGML_TYPE_Q8_1:
default:
{
GGML_ASSERT(false);