diff --git a/convert-llama-h5-to-gguf.py b/convert-llama-h5-to-gguf.py
index 56063f375..35f85bb97 100644
--- a/convert-llama-h5-to-gguf.py
+++ b/convert-llama-h5-to-gguf.py
@@ -327,7 +327,7 @@ for part_name in part_names:
         if ftype == 1 and data_dtype == np.float32 and name.endswith(".weight") and n_dims == 2:
             data = data.astype(np.float16)
 
-        print( name + ", shape " + str(len(data.shape)) + ", " + str(old_dtype) + " --> " + str(data.dtype))
+        print(name + ", shape " + str(len(data.shape)) + ", " + str(old_dtype) + " --> " + str(data.dtype))
 
         gguf_writer.write_tensor_to_file(data)
 
diff --git a/ggml.c b/ggml.c
index 19fbe6b9e..f7e72ed84 100644
--- a/ggml.c
+++ b/ggml.c
@@ -9140,6 +9140,8 @@ static void ggml_compute_forward_mul(
         const struct ggml_tensor * src0,
         const struct ggml_tensor * src1,
         struct ggml_tensor * dst) {
+    GGML_ASSERT(src1->type == GGML_TYPE_F32 && "only f32 src1 supported for now");
+
     switch (src0->type) {
         case GGML_TYPE_F32:
             {