I can't find 204,045 data for training.
I load this dataset and check the data,I found
dataset = datasets.load_dataset("GEM/xsum")
dataset
DatasetDict({
train: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 23206
})
validation: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 1117
})
test: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 1166
})
challenge_train_sample: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 500
})
challenge_validation_sample: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 500
})
challenge_test_backtranslation: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 500
})
challenge_test_bfp_02: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 500
})
challenge_test_bfp_05: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 500
})
challenge_test_nopunc: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 500
})
challenge_test_covid: Dataset({
features: ['gem_id', 'gem_parent_id', 'xsum_id', 'document', 'target', 'references'],
num_rows: 401
})
})