Index not found when using ALS algorithm with Implicit library

Question

I am trying to create a recommendation system with Python and a very small data, in this case, I have more users than items (SD)

    SD                     user    value
29  Hutsa kafeinagabea  jolaizolaa  32
23  Hurradun Capuccinoa ibaraia 117
37  Kafe Hutsa  ivelez  1
21  Hurradun Capuccinoa aalonso 7
67  Txokolatea  dreguera    2
18  Ebakia kafeinagabea dreguera    1
44  Kafe Luzea  ezugasti    17
17  Ebakia  xesnaola    67
71  Txokolatea  ivelez  88
40  Kafe Hutsa  miturbe 1

I create the sparse matrix using the data of my data set

grouped_df['SD'] = grouped_df['SD'].astype("category")
grouped_df['user'] = grouped_df['user'].astype("category")

grouped_df['SD_id'] = grouped_df['SD'].cat.codes
grouped_df['user_id'] = grouped_df['user'].cat.codes
sparse_content_person = sparse.csr_matrix((grouped_df['value'].astype(float), (grouped_df['SD_id'], grouped_df['user_id'])))

then I train my model

model = implicit.als.AlternatingLeastSquares(factors=10, regularization=0.1, iterations=50)
alpha = 1
data = (sparse_content_person * alpha).astype('double')
model.fit(data)

Then I try to get the items that are similar to code/index 11

model.similar_items(11)

and I get the following result:

(array([11,  2, 19,  9,  3, 23, 20,  6, 13, 24], dtype=int32),
 array([1.        , 1.        , 0.9999995 , 0.8310417 , 0.8063449 ,
        0.7573321 , 0.6772214 , 0.6740392 , 0.4807193 , 0.47637755],
       dtype=float32))

These indices don't exist as item codes, since I only have 16 items. I could get the product index from the sparse_content_person matrix, but I don't know if I am correct. I am following a tutorial and they get the info directly from the dataframe (tutorial link: https://towardsdatascience.com/building-a-collaborative-filtering-recommender-system-with-clickstream-data-dffc86c8c65).