Upload README.md with huggingface_hub

README.md

@@ -1,25 +1,38 @@
 ---
-dataset_info:
-  features:
-  - name: question
-    dtype: string
-  - name: target
-    dtype: string
-  - name: attack
-    dtype: string
-  - name: adv_prompt
-    dtype: string
-  - name: best_loss
-    dtype: float64
-  splits:
-  - name: train
-    num_bytes: 242348
-    num_examples: 550
-  download_size: 150183
-  dataset_size: 242348
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
+license: mit
+task_categories:
+- text-classification
+language:
+- en
+tags:
+- safety
+- jailbreak
+- gcg
+- adversarial
+- llm
 ---
+
+# harmeval-gcg-llama3-1b
+
+HarmEval ([SoftMINER-Group/HarmEval](https://huggingface.co/datasets/SoftMINER-Group/HarmEval)) augmented with adversarial suffixes generated via the **Greedy Coordinate Gradient (GCG)** attack method, optimized specifically against **Llama-3.2-1B-Instruct**.
+
+Each harmful prompt is paired with a GCG-optimized adversarial suffix that, when appended to the original question, maximizes the probability of the model producing a target harmful response.
+
+## Columns
+
+| Column | Description |
+|---|---|
+| `question` | Original harmful prompt from HarmEval |
+| `target` | Target response prefix used during GCG optimization (e.g. `"Sure, here is the answer:"`) |
+| `attack` | Raw GCG adversarial suffix (gibberish token sequence) |
+| `adv_prompt` | Full adversarial prompt: `question + attack` (ready to use as model input) |
+| `best_loss` | Cross-entropy loss of the target at the end of optimization (lower = stronger attack) |
+
+## Statistics
+
+- **550 rows** (one per HarmEval question)
+- Suffix search run for a fixed budget of GCG steps targeting Llama-3.2-1B-Instruct
+
+## Intended use
+
+Evaluating the robustness of safety classifiers (e.g. linear probes, GLP diffusion models) against GCG-style adversarial inputs at the activation level.