Upload OpenELM-Safety-LoRA v8 adapter

README.md

@@ -117,7 +117,7 @@ The model expects prompts formatted with a `<|safety|>` prefix:
 
 ## Training Data
 
-The model was fine-tuned on a curated dataset of ~3,000 examples:
+The model was fine-tuned on a curated dataset of ~3,000 examples, available at [ApdoElepe/openelm-safety-dataset](https://huggingface.co/datasets/ApdoElepe/openelm-safety-dataset):
 
 | Type | Count | Source |
 |------|-------|--------|