datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
wikipedia-summary-dataset-128k
Wikipedia Summary Dataset 128k
This is random subsample of 128k entries from the wikipedia summary dataset, processed with the following code:
import pandas as pd
df = pd.read_parquet('wikipedia-summary.parquet')
df['l'] = df['summary'].str.len()
rdf = df[(df['l'] > 300) & (df['l'] < 600)]
# Filter out any rows 'topic' that have non-alphanumeric characters
mask = rdf['topic'].str.contains(r'^[a-zA-Z0-9 ]+$') == True
rdf = rdf[mask == True].sample(128000)[['topic'… See the full description on the dataset page: https://huggingface.co/datasets/54rt1n/wikipedia-summary-dataset-128k.Phi_3_mini_128k_mathPhi_3_mini_128k_CSPhi_3_mini_128k_biology
Phi_3_mini_128k_biology
license: mit
Phi_3_mini_128k_physicsmistral_128k