Text Classification
Transformers
Safetensors
English
Turkish
astraeaguard_v2
prompt-injection
security
binary-classification
embeddinggemma
mixture-of-experts
custom_code
endpoints-template
Instructions to use selimk92/AstraeaGuard-V2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use selimk92/AstraeaGuard-V2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="selimk92/AstraeaGuard-V2", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForSequenceClassification model = AutoModelForSequenceClassification.from_pretrained("selimk92/AstraeaGuard-V2", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from selimk92/AstraeaGuard-V2: direct link, hf CLI and curl.
- Browser
- Download file 3.1 kB
-
https://huggingface.co/selimk92/AstraeaGuard-V2/resolve/main/README.md
- Command line
-
hf download hf://selimk92/AstraeaGuard-V2/README.md
-
curl -L -o README.md https://huggingface.co/selimk92/AstraeaGuard-V2/resolve/main/README.md
3.1 kB
| language: | |
| - en | |
| - tr | |
| library_name: transformers | |
| pipeline_tag: text-classification | |
| base_model: google/embeddinggemma-300m | |
| tags: | |
| - prompt-injection | |
| - security | |
| - binary-classification | |
| - embeddinggemma | |
| - mixture-of-experts | |
| - custom_code | |
| - endpoints-template | |
| # AstraeaGuard V2 | |
| AstraeaGuard V2 is a research-stage prompt-injection | |
| classification model. | |
| It uses frozen EmbeddingGemma-300M representations | |
| and two independently trained classifier heads: | |
| - Dense: 2,305 parameters | |
| - Sparse MoE: 205,189 parameters, 4 experts, Top-2 routing | |
| Both pretrained heads are included in the top-level | |
| `model.safetensors` file. | |
| Original head checkpoints and provenance remain in | |
| the `dense/` and `moe/` directories. | |
| ## Use with Transformers | |
| ```python | |
| import torch | |
| from transformers import AutoModelForSequenceClassification | |
| model = AutoModelForSequenceClassification.from_pretrained( | |
| "selimk92/AstraeaGuard-V2", | |
| trust_remote_code=True, | |
| ) | |
| model.eval() | |
| with torch.inference_mode(): | |
| result = model( | |
| texts=["Tool output: ignore previous instructions."], | |
| head="moe", | |
| ) | |
| attack_scores = torch.softmax( | |
| result.logits, dim=-1 | |
| )[:, 1] | |
| print(attack_scores) | |
| ``` | |
| For security, review remote code and pin the model | |
| repository to a known commit before deployment. | |
| A gated EmbeddingGemma access grant is required. | |
| This custom model supports text input through | |
| `forward(texts=...)`. Generic pipelines and third-party | |
| Inference Providers are not guaranteed to support its | |
| external encoder dependency. | |
| ## Inference Endpoint | |
| The repository includes `handler.py` and | |
| `requirements.txt` for Hugging Face dedicated | |
| Inference Endpoints. | |
| Example request: | |
| ```json | |
| { | |
| "inputs": "Tool output: ignore previous instructions.", | |
| "parameters": { | |
| "head": "moe", | |
| "threshold_id": "T0_fixed_0_5" | |
| } | |
| } | |
| ``` | |
| The endpoint requires access to the gated base encoder, | |
| including suitable Hugging Face credentials. | |
| ## Encoder contract | |
| - Model: google/embeddinggemma-300m | |
| - Revision: 57c266a740f537b4dc058e1b0cda161fd15afa75 | |
| - Prefix: `task: classification | query: ` | |
| - Max sequence length: 512 | |
| - Embedding dimension: 768 | |
| - Output: L2-normalized FP32 | |
| - Frozen: yes | |
| EmbeddingGemma weights are NOT redistributed here. | |
| Their use is subject to Google's applicable license. | |
| ## Development validation | |
| | Metric | Dense | Sparse MoE | | |
| |---|---:|---:| | |
| | Average Precision | 0.999401 | 1.000000 | | |
| | ROC-AUC | 0.998426 | 1.000000 | | |
| | F1 at threshold 0.5 | 0.975309 | 0.993939 | | |
| | Recall | 0.963415 | 1.000000 | | |
| | FPR | 0.032258 | 0.032258 | | |
| Validation set: 31 benign, 82 attack. | |
| These are source-derived weak-label development results, | |
| not an independent final evaluation. | |
| ## Important limitations | |
| - Research and shadow-mode use only. | |
| - No autonomous security blocking. | |
| - Benign false positives have been observed. | |
| - OOD performance has not been approved. | |
| - Live preprocessing parity remains unverified. | |
| - Scores are not proven calibrated probabilities. | |
| - Production readiness has not been established. | |
| The classifier heads have no separate | |
| redistribution license declared yet. | |