Unicode Decode Error in nonneural_models.ipynb

[KataTech]

Hi! I am attempting to explore the dataset and code snippet associated with the Reddit Community Interaction and Conflicts paper. When attempting to reproduce results from the nonneural_models.ipynb script, I encountered the following error trying to load in the user, source, and target community embeddings.

The same issue occurs for loading the post embeddings for LSTM. I had no problems loading in the handcrafted features, so it is likely not a problem with path or downloading the wrong data, but it may deal with outdated packages or decompression of the zip file? Please let me know if there are any suggestions!