Why is the prediction length longer than expected with this LSTM model?

Question

I'm working on a multivariate time series forecasting problem where I have 2 features and 15 time steps and I want to predict one/two future values. My data set shapes are as follows:

Training Shape: (2063, 15, 2) (2063, 1)
Testing Shape: (259, 15, 2) (259, 1)
Prediction Shape: (65, 15, 2) (65, 1)

I have the following LSTM model:

def mylstm(structure , data):
    inputs = Input(shape=(data["X_train"].shape[1],data["X_train"].shape[2]))
    lstm_1 = LSTM(structure["n_hidden_units"], return_sequences=True,
                  activation=structure["activation"],
                  kernel_initializer=structure["network_weight_initial"])(inputs)
    lstm_out=LSTM(structure["n_hidden_units"])(lstm_1)
    outputs = Dense(1)(lstm_out)
    model = Model(inputs=inputs, outputs=outputs)
    model.summary()
    return model

optimizer = getattr(optimizers, structure["opt"])(learning_rate=structure["learning_rate"])
model.compile(optimizer=optimizer, loss='mse')
history = model.fit(
        data["X_train"], data["y_train"], epochs=structure["epoch"], 
        batch_size=structure["batch_size"], verbose=2,
        validation_data=(X_valid, y_valid), shuffle=False, 
        )

After training the model and calling the predict function as follows

y_pred = model.predict(data["X_test"])

I get the error message:

ValueError: y_true and y_pred must have the same number of dimensions.

That's because y_pred is 1820 length (65*14*2). I cannot figure out why it considers the batch size, I was supposed to get 65*2 dimensions but it gives 1820*1.