Upload dataset card

README.md

@@ -1,19 +1,58 @@
 ---
-dataset_info:
-  features:
-  - name: text
-    dtype: string
-  - name: slug
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 3432495
-    num_examples: 50
-  download_size: 1971687
-  dataset_size: 3432495
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
+language:
+- en
+license: cc-by-sa-4.0
+size_categories:
+- n<1K
+source_datasets:
+- original
+task_categories:
+- text-generation
+pretty_name: Top US Cities (Wikipedia)
+tags:
+- geography
+- wikipedia
+- usa
+- plain-text
+- cities
+- census
 ---
+
+# Dataset Card for Top US Cities (Wikipedia)
+
+Wikipedia plain-text extracts for the top 50 US cities by 2020 Census population (Washington, D.C. omitted).
+
+## Dataset Details
+
+- **Language:** English
+- **License:** [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/) (Wikipedia text)
+- **Source:** English Wikipedia (plain-text extracts)
+- **Rows:** 50
+
+## Dataset Description
+
+One row per city with `text` (full English Wikipedia plain-text extract) and `slug` (filesystem-safe place identifier). City selection follows 2020 Census population ranking. Useful for language modeling, retrieval, and geography-focused NLP.
+
+## Dataset Structure
+
+One row per place. Single split: `train`.
+
+| Field | Type | Description |
+|-------|------|-------------|
+| `text` | string | Full Wikipedia plain-text extract for the place |
+| `slug` | string | Filesystem-safe place identifier |
+
+
+### Example
+
+```python
+from datasets import load_dataset
+
+ds = load_dataset("codycollier/geo-us-cities-top")
+row = ds["train"][0]
+print(row["slug"], len(row["text"]))
+```
+
+## Licensing Information
+
+Text content originates from Wikipedia and remains under the Creative Commons Attribution-ShareAlike License. See [Wikipedia:Copyrights](https://en.wikipedia.org/wiki/Wikipedia:Copyrights).