A small set of 2048 samples from HuggingFaceH4/ultrachat_200k for easy calibration.
DATASET_ID = "HuggingFaceH4/ultrachat_200k"
DATASET_SPLIT = "train_sft"
SAMPLE_SIZE = 2048
NEW_DATASET_ID = "mgoin/ultrachat_2k"
sampled_ds = load_dataset(DATASET_ID, split=DATASET_SPLIT).shuffle(seed=42).select(range(SAMPLE_SIZE))… See the full description on the dataset page:
https://huggingface.co/datasets/mgoin/ultrachat_2k.