Is additional validation set affect training result?

Training with following setting:
tree_metho: “gpu_hist”,
objective: “reg:squarederror”
eval_metric: [“rmse”]

watchlist=[(dtrain, “train”)]
model_fiitted = xgb.train(
params,
dtrain,
evals=watchlist,
)

Will training result be different by replacing watchlist with [(dtest, “test”), (dtrain, “train”)], dose the additional dtest has any impact on training process?