Text Classification
Transformers
Safetensors
English
sentinel_stage_a
feature-extraction
custom
compliance
finance
risk-detection
sentinel-stage-a
limited-functionality
model-version:sentinel-mb-c-d11-20260424
custom_code
Instructions to use AurelexAI/sentinel-01-pub with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AurelexAI/sentinel-01-pub with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="AurelexAI/sentinel-01-pub", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AurelexAI/sentinel-01-pub", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update sentinel-mb-c-d11 model card
Browse files
README.md
CHANGED
|
@@ -39,11 +39,14 @@ This repository is intended to publish only the designated limited-functionality
|
|
| 39 |
|
| 40 |
- Repo ID: `AurelexAI/sentinel-01-pub`
|
| 41 |
- Model key: `sentinel-mb-c-d11`
|
| 42 |
-
-
|
| 43 |
- Release channel: `sentinel-01-pub`
|
|
|
|
| 44 |
- Artifact format: `transformers_end_to_end`
|
| 45 |
- Publication status: public, approved by Aurelex on 2026-04-28
|
| 46 |
|
|
|
|
|
|
|
| 47 |
The model was selected as a public, lower-capacity, limited-functionality variant. It is separate from Aurelex production channels and full-featured internal models.
|
| 48 |
|
| 49 |
## Loading From Hugging Face
|
|
@@ -71,7 +74,9 @@ print(result)
|
|
| 71 |
print(model_version)
|
| 72 |
```
|
| 73 |
|
| 74 |
-
|
|
|
|
|
|
|
| 75 |
|
| 76 |
## Outputs
|
| 77 |
|
|
@@ -81,7 +86,7 @@ These outputs are risk-review signals for human review. They are not final compl
|
|
| 81 |
|
| 82 |
## Evaluation
|
| 83 |
|
| 84 |
-
Dataset: `2026-04-07-final-audit-clear-v1`.
|
| 85 |
|
| 86 |
| Metric | Test |
|
| 87 |
| --- | ---: |
|
|
|
|
| 39 |
|
| 40 |
- Repo ID: `AurelexAI/sentinel-01-pub`
|
| 41 |
- Model key: `sentinel-mb-c-d11`
|
| 42 |
+
- Current model artifact version: `sentinel-mb-c-d11-20260424`
|
| 43 |
- Release channel: `sentinel-01-pub`
|
| 44 |
+
- Source encoder lineage: `answerdotai/ModernBERT-base`
|
| 45 |
- Artifact format: `transformers_end_to_end`
|
| 46 |
- Publication status: public, approved by Aurelex on 2026-04-28
|
| 47 |
|
| 48 |
+
The current public release is the `sentinel-01-pub` channel serving the `sentinel-mb-c-d11-20260424` artifact. The repository ships as a self-contained end-to-end model; the source encoder lineage is listed for transparency and Hugging Face discoverability, not because users need to load a separate encoder.
|
| 49 |
+
|
| 50 |
The model was selected as a public, lower-capacity, limited-functionality variant. It is separate from Aurelex production channels and full-featured internal models.
|
| 51 |
|
| 52 |
## Loading From Hugging Face
|
|
|
|
| 74 |
print(model_version)
|
| 75 |
```
|
| 76 |
|
| 77 |
+
`trust_remote_code=True` is required because this repository includes custom Transformers runtime files: `configuration_sentinel.py`, `modeling_sentinel.py`, and `pipeline_sentinel.py`. Those files register the `sentinel-stage-a` pipeline and the custom Sentinel model class used to interpret the serialized weights.
|
| 78 |
+
|
| 79 |
+
Only use `trust_remote_code=True` for repositories you trust and whose code you have reviewed. For reproducible use, pin a reviewed Hub commit with `revision="<commit_sha>"` so the loaded runtime code and weights cannot change unexpectedly.
|
| 80 |
|
| 81 |
## Outputs
|
| 82 |
|
|
|
|
| 86 |
|
| 87 |
## Evaluation
|
| 88 |
|
| 89 |
+
Dataset: `2026-04-07-final-audit-clear-v1`, test split size `150`.
|
| 90 |
|
| 91 |
| Metric | Test |
|
| 92 |
| --- | ---: |
|