Text Classification
Transformers
Safetensors
modernbert
ai-safety
safeguards
guardrails
text-embeddings-inference
Instructions to use dcarpintero/pangolin-guard-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use dcarpintero/pangolin-guard-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="dcarpintero/pangolin-guard-base")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("dcarpintero/pangolin-guard-base") model = AutoModelForSequenceClassification.from_pretrained("dcarpintero/pangolin-guard-base", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -22,9 +22,7 @@ PangolinGuard is a ModernBERT (Base), lightweight model that discriminates malic
|
|
| 22 |
|
| 23 |
## Intended uses
|
| 24 |
|
| 25 |
-
- Adding
|
| 26 |
-
- Topic and content moderation
|
| 27 |
-
- Mitigating risks when connecting AI pipelines to external services
|
| 28 |
|
| 29 |
## Evaluation data
|
| 30 |
|
|
|
|
| 22 |
|
| 23 |
## Intended uses
|
| 24 |
|
| 25 |
+
- Adding a self-hosted, inexpensive safety checks (against prompt injection attacks) to AI agents and conversational interfaces.
|
|
|
|
|
|
|
| 26 |
|
| 27 |
## Evaluation data
|
| 28 |
|