Text Classification
Transformers
Safetensors
English
modernbert
refusal-classification
refusal-detection
ai-safety
ai-alignment
heretic
text-embeddings-inference
Instructions to use saidutta69/vantablack with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use saidutta69/vantablack with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="saidutta69/vantablack")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("saidutta69/vantablack") model = AutoModelForSequenceClassification.from_pretrained("saidutta69/vantablack", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
Yesterday
-
Card: link the reporting discussion
Verified signatureMislabelled examples and integration bugs go to a thread with a template, so reports arrive reproducible instead of freeform.c0aae29 -
Card: state the 512-token ceiling and where it does and does not matter
Verified signatureHeretic scores a prompt plus about 100 generated tokens, so 512 is enough there. For general use the response tail is invisible to the model, which the card now says instead of leaving it to the Usage note.19625d9 -
Card figure: stop the confusion-matrix x-axis labels colliding
Verified signatureShort x tick labels; the y axis keeps the full class names.c0341a4 -
Card: correct the Brier comparison, name the four residual errors, pin the Heretic sha
Verified signatureBrier 0.0173 is 40% below v2's 0.0287, not 'under half'. The four test errors are now described as they are. The Heretic snippet carries the real commit instead of a placeholder.980b949
2 days ago
-
Model card: comparison as charts, no dataset version label
Verified signatureReplaces the results tables with charts on a consistent basis: baselines at their calibrated behaviour, VantaBlack at its baked-in calibration, so all three are on the same footing. Adds a per-class comparison and a training curve with the selected epoch marked. Drops the two auto-generated figures that were unreadable at this value range. Removes the dataset version from the card; the version lives on the dataset.d3caa9c -
VantaBlack on dataset v5: calibrated ModernBERT-large, macro F1 0.9911
Verified signatureWeights are the s17 checkpoint trained on refusal-classification-dataset v5, with validation-fitted calibration (T=0.75, compliant bias 1.0) folded into the classifier head. Plain argmax reproduces the reported predictions exactly: 445 of 445 test rows identical, 4 errors. Replaces the 0.7924 checkpoint, and removes figures and provenance files from that older model that no longer describe these weights.dcf6697
12 days ago
-
Calibration is now baked into the published weights (see Calibration section)
Verified signaturef45165c -
Provenance: calibration folded into the head
Verified signature0f53417 -
Val predictions of the shipped (fused) checkpoint
Verified signature5a730ab -
Test predictions of the shipped (fused) checkpoint
Verified signature52ff5f9 -
Metrics of the shipped (fused) checkpoint
Verified signature693acac -
tokenizer for fused checkpoint
Verified signature385d3a3 -
tokenizer for fused checkpoint
Verified signatureabc8d6e -
Swap weights: fold validation calibration into classifier head
Verified signature09b252b -
Swap weights: fold validation calibration into classifier head
Verified signature498323b -
Use the stat strip for the headline numbers
Verified signature5f5c441 -
Add branded stat strip
Verified signature973f2e7 -
Drop in-image titles; keep section headings on the card
Verified signature8779963 -
Drop in-image titles; keep section headings on the card
Verified signature0cda831 -
Drop in-image titles; keep section headings on the card
Verified signaturebdbfef6 -
Drop in-image titles; keep section headings on the card
Verified signature438bf10 -
Clarify teacher labels
Verified signaturecc9e129 -
Rebuild model card in branded layout with dark figures
Verified signature9401834 -
Restyle seed-spread figure in brand palette
Verified signature8c119f8 -
Restyle training figure in brand palette
Verified signature1868a16 -
Restyle per-class figure in brand palette
Verified signature4c2109e -
Restyle headline figure in brand palette
Verified signatureb3b43a9 -
Add branded banner
Verified signature64d464a -
Model card: add training curve
Verified signaturec175c38 -
Add training loss and validation F1 figure
Verified signaturefcff9bd -
Model card: add headline, seed-spread, and per-class figures
Verified signature171c1c8 -
Add model card figure: seeds
Verified signatureaf8330e -
Add model card figure: per_class
Verified signaturec51a13b -
Add model card figure: headline
Verified signature979646d -
Add model card
Verified signaturefb6d2d9 -
VantaBlack candidate: ModernBERT-large s17, 0.792 macro F1 on clean 321-row test
Verified signature74fc3f8 -
initial commit
Verified signature806ed2a