diff --git a/docs/config.qmd b/docs/config.qmd
index caa9b7649f..dadc5c487c 100644
--- a/docs/config.qmd
+++ b/docs/config.qmd
@@ -268,6 +268,7 @@ torch_compile_backend:  # Optional[str]
 # If greater than 1, backpropagation will be skipped and the gradients will be accumulated for the given number of steps.
 gradient_accumulation_steps: 1
 # The number of samples to include in each batch. This is the number of samples sent to each GPU.
+# Batch size per gpu = micro_batch_size * gradient_accumulation_steps
 micro_batch_size: 2
 eval_batch_size:
 num_epochs: 4