How to use lhallee/speedrun with Transformers:
# Load model directly from transformers import PLM model = PLM.from_pretrained("lhallee/speedrun", dtype="auto")
