make it compatible with dataset preview

README.md

@@ -8,7 +8,12 @@ tags:
 - files
 - epstein-files
 size_categories:
-- 1M<n<10M
+- 100G<n<1T
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: "*.parquet"
 ---
 
 # The Epstein Files Dataset
@@ -24,20 +29,13 @@ The dataset is presented in processed formats, including extracted text from PDF
 
 ## Dataset Creation
 
-### Curation Rationale
-The primary goal of this dataset is to enhance accessibility to public domain Epstein-related documents for researchers, journalists, and the general public, consolidating dispersed resources into a single, user-friendly repository on Hugging Face.
-
-### Source Data
-
-#### Initial Data Collection or Creation
-The data originates from a public torrent containing official releases. To ensure transparency and reproducibility, the raw sources align with government archives like the DOJ Epstein Library and FBI Vault.
-
 ### Processing Details
+The data is stored in Apache Parquet format.
 - **Text Extraction**: Automated extraction of text from PDF documents.
-- **Multimedia Handling**: Audio, images, and videos extracted and stored as binary files optimized for Hugging Face datasets.
+- **Multimedia Handling**: Audio, images, and videos are stored as binary blobs within the parquet structs.
 
 ### Recreate the Dataset
-To recreate this dataset exactly as provided, refer to the GitHub repository at https://github.com/Nikityyy/Epstein-Files. The repository includes scripts and instructions for downloading the source torrent, performing the processing steps ), and preparing the data for upload to Hugging Face.
+To recreate this dataset exactly as provided, refer to the GitHub repository at https://github.com/Nikityyy/Epstein-Files.
 
 ## Additional Information