data_prep.py

import json
from pathlib import Path

def chunk_text(text: str, chunk_size=500, overlap=50):
    """Split text into overlapping word chunks."""
    words = text.split()
    for i in range(0, len(words), chunk_size - overlap):
        yield " ".join(words[i:i+chunk_size])

def split_to_jsonl(input_file: str, output_file: str = "pretrain.jsonl",
                   chunk_size=500, overlap=50):
    """Split raw file into JSONL for pretraining (no cleaning)."""
    raw_text = Path(input_file).read_text(encoding="utf-8")

    with open(output_file, "w", encoding="utf-8") as f:
        for chunk in chunk_text(raw_text, chunk_size, overlap):
            record = {"text": chunk}
            f.write(json.dumps(record, ensure_ascii=False) + "\n")

    print(f"✅ Saved dataset → {output_file}")

if __name__ == "__main__":
    # Example usage
    input_file = "gist.md"   # your markdown/text file
    split_to_jsonl(input_file, "pretrain.jsonl", chunk_size=500, overlap=50)