Update fn name

EricLBuehler · Mar 12, 2024 · 9140e50 · 9140e50
1 parent 832540f
commit 9140e50
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/candle-kernels/src/layernorm_kernels.cu b/candle-kernels/src/layernorm_kernels.cu
@@ -77,7 +77,7 @@ extern "C" __global__ void rms_norm_f64(
 
 #if __CUDA_ARCH__ >= 800
 #include <cuda_bf16.h>
-extern "C" __global__ void rms_norm_f64(
+extern "C" __global__ void rms_norm_bf16(
     __nv_bfloat16 *__restrict__ out, \
     const __nv_bfloat16 *__restrict__ input, \
     const double epsilon, const uint32_t num_tokens, \