{
    "dataset_repo": "EleutherAI/fineweb-edu-dedup-10b",
    "dataset_split": "train",
    "dataset_name": "",
    "dataset_column": "text",
    "batch_size": 16,
    "cache_ctx_len": 256,
    "n_tokens": 1000000,
    "n_splits": 5,
    "model_name": "meta-llama/Llama-3.2-1B"
}