Common Pile v0.1 Filtered Data Collection An LLM pre-training dataset produced by filtering and deduplicating the raw text collected in the Common Pile v0.1 • 31 items • Updated Jun 6 • 20