Upload README.md with huggingface_hub

README.md

@@ -1,23 +1,30 @@
 ---
-dataset_info:
-  features:
-  - name: id
-    dtype: string
-  - name: section_text
-    dtype: string
-  - name: label
-    dtype: string
-  - name: is_synthetic
-    dtype: bool
-  splits:
-  - name: train
-    num_bytes: 198096
-    num_examples: 200
-  download_size: 98243
-  dataset_size: 198096
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
+tags:
+- classify-and-augment
+- llm-annotated
 ---
+
+# davanstrien/otb-augmented
+
+LLM-annotated dataset produced by [classify-and-augment](https://github.com/davanstrien/classify-and-augment).
+
+## Configuration
+
+- **Model**: `HuggingFaceTB/SmolLM3-3B`
+- **Labels**: `jim_crow`, `no_jim_crow`
+- **Input rows**: 200
+- **Output rows**: 200
+
+## Label distribution
+
+| Label | Real | Synthetic | Total |
+|---|---:|---:|---:|
+| `no_jim_crow` | 200 | 0 | 200 |
+
+## Synthesis audit
+
+| Class | Needed | Generated | Validated | Kept | Acceptance |
+|---|---:|---:|---:|---:|---:|
+| `jim_crow` | 50 | 0 | 0 | 0 | 0.0% |
+
+Acceptance = synthetic candidates that the same model re-classified as the target class (self-consistency check, [Synthetic Imputation, arxiv 2504.15160](https://arxiv.org/abs/2504.15160)).