# Step 1: Load dataset and model tokenizer
dataset = load_dataset('imdb')
tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')