Upload README.md with huggingface_hub

README.md

@@ -2,35 +2,6 @@
 language: en
 task_categories:
 - text-generation
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-dataset_info:
-  features:
-  - name: example_id
-    dtype: int64
-  - name: metadata
-    dtype: string
-  - name: classification_prompt
-    dtype: string
-  - name: classification_completion
-    dtype: string
-  - name: classification_text
-    dtype: string
-  - name: improved_signature
-    dtype: string
-  - name: improved_model_weights
-    dtype: string
-  - name: training_metrics
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 265157953
-    num_examples: 8300
-  download_size: 122090368
-  dataset_size: 265157953
 ---
 # Subject Models for Interpretability Training
 
@@ -67,7 +38,7 @@ These examples are intended for training an interpreter to:
 
 | Task Type | Min Tokens | Max Tokens | Avg Tokens |
 |-----------|------------|------------|------------|
-| Classification | 2628 | 6519 | 4315.3 |
+| Classification | 2628 | 6519 | 4316.2 |
 
 
 ## Dataset Fields