nemo_rl.data.datasets.preference_datasets.helpsteer3#

Module Contents#

Classes#

HelpSteer3Dataset

HelpSteer3 preference dataset for DPO training.

API#

class nemo_rl.data.datasets.preference_datasets.helpsteer3.HelpSteer3Dataset(split: str = 'train', **kwargs)#

Bases: nemo_rl.data.datasets.raw_dataset.RawDataset

HelpSteer3 preference dataset for DPO training.

Parameters:

split – Split name for the dataset, default is “train”

Initialization

format_data(data: dict[str, Any]) dict[str, Any]#