Is there a way to get the vocabulary size of an initialized tokenizer from `var encoding = GptEncoding.GetEncoding(encodingName);`?