Instructions to use arthrod/c3750-mdeberta_base-vanilla-1-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- GLiNER
How to use arthrod/c3750-mdeberta_base-vanilla-1-2 with GLiNER:
from gliner import GLiNER model = GLiNER.from_pretrained("arthrod/c3750-mdeberta_base-vanilla-1-2") - Notebooks
- Google Colab
- Kaggle
c3750-mdeberta_base-vanilla-1-2
Legacy "nightmare" PII run on microsoft/mdeberta-v3-base with span/marker head. Multiple checkpoints up to step 3750.
- Version: checkpoints (250, 1000 … 3750)
- Backbone: mDeBERTa-v3-base
- GLiNER mode: vanilla / markerV0
- Weights per ckpt: ~1.15 GB
Training
- Dataset: legacy multi-PII experiment (see
arthrod/v1-mdeberta_base-vanilla-1-1for the clean v1 release)
Evaluation
Not benchmarked — legacy experimental run.
- Downloads last month
- 1