diff --git a/megatron/core/tensor_parallel/layers.py b/megatron/core/tensor_parallel/layers.py
index 2245113c9c..7f1d217ac0 100644
--- a/megatron/core/tensor_parallel/layers.py
+++ b/megatron/core/tensor_parallel/layers.py
@@ -368,7 +368,7 @@ def backward(ctx, grad_output):
         #     grad_weight = grad_output.t().matmul(total_input)
         from megatron.core.tensor_parallel.weight_grad_store import WeightGradStore
         WeightGradStore.put(total_input, grad_output, weight, gradientUpdateFunction)
-        grad_weight = None
+        grad_weight = weight.grad
         grad_bias = grad_output.sum(dim=0) if use_bias else None
 
         if ctx.sequence_parallel: