ggml : fix I8MM Q4_1 scaling factor conversion (#10562)

ggml-ci
2024-11-29 16:25:39 +02:00 · 2024-11-29 16:25:39 +02:00 · f0678c5ff4
commit f0678c5ff4
parent 4b3242bbea
3 changed files with 38 additions and 25 deletions
--- a/tests/test-backend-ops.cpp
+++ b/tests/test-backend-ops.cpp
@ -3334,7 +3334,9 @@ static const ggml_type all_types[] = {

 static const ggml_type base_types[] = {
    GGML_TYPE_F32, GGML_TYPE_F16,
+    GGML_TYPE_Q8_0, // for I8MM tests
    GGML_TYPE_Q4_0,
+    GGML_TYPE_Q4_1, // for I8MM tests
    GGML_TYPE_Q4_K,
    GGML_TYPE_IQ2_XXS
 };