args:
{'add_timestamp': True,
'dataset_end_idx': 37386,
'dataset_mixer_list': ['VGraf/tulu_paraphrase_questions_BIG_70000', '1.0'],
'dataset_shuffle_seed': -1,
'dataset_splits': ['train', 'train'],
'dataset_start_idx': 0,
'hf_entity': 'VGraf',
'hf_repo_id': 'generation',
'max_qs': 8,
'mode': 'generation'… See the full description on the dataset page:
https://huggingface.co/datasets/VGraf/generation_1746587442.
