import seaborn as sns
import matplotlib.pyplot as plt

df_test = pd.read_csv(DATA_PATH + "test.csv")
df_test = preprocess(df_test)

test_token_length = tokenizer(
    text=df_test["comment_text"].tolist(), return_length=True
).length

plt.figure(figsize=(20, 6))
sns.distplot(test_token_length, bins=100)
plt.xticks(np.arange(0, 400, 50))
plt.show()