ValueError: A KerasTensor is symbolic: it's a placeholder for a shape an a dtype. It doesn't have any actual numerical value

Question

I am getting below error ValueError: A KerasTensor is symbolic: it's a placeholder for a shape an a dtype. It doesn't have any actual numerical value. You cannot convert it to a NumPy array.

for the code

def build_model():
    # Create a DebertaV3Classifier model
    classifier = keras_nlp.models.DebertaV3Classifier.from_preset(
        'deberta_v3_base_en',
        preprocessor=None,
        num_classes=2 # one output per one option, for five options total 5 outputs
    )
    inputs = classifier.input
    logits = classifier(inputs)
    print(inputs)
    print(logits)

    # Compute final output
    

    outputs = keras.layers.Activation('sigmoid',name="sigmoid")(logits)
    model = keras.Model(inputs, outputs)

    # Compile the model with optimizer, loss, and metrics
    model.compile(
        optimizer=keras.optimizers.AdamW(5e-6),
        loss=keras.losses.BinaryCrossentropy(label_smoothing=0.02),
        metrics=[
            keras.metrics.AUC(name="auc"),
        ],
        jit_compile=True
    )
    return model

Same code is working on kaggle notebook without any issue but when running on colab then error