Add data_pipeline.py — tokenize_user_sequences, pack_sequences, prepare_clm_dataset

1dfd4e2

verified

rtferraz commited on 8 days ago