metadata

library_name: transformers
license: apache-2.0
datasets:
  - Mozilla/smart_ner_dataset
language:
  - en
base_model:
  - distilbert/distilbert-base-uncased
pipeline_tag: token-classification

Fine-tuned LoRA Token classification on distilbert

This is a fine-tuned LoRA token classifier on distilbert, designed for NER on multiple categories PERSON, ORG, CITY, STATE, CITY_STATE.

Model Details

Model Description

This model is based on distilbert/distilbert-base-uncased and fine-tuned using LoRA for token classification. The fine-tuning process adapts the model to predict tokens across 10 categories:

"O"  # Outside any named entity
"B-PER"       # Beginning of a person entity
"I-PER"       # Inside a person entity
"B-ORG"       # Beginning of an organization entity
"I-ORG"       # Inside an organization entity
"B-CITY"      # Beginning of a city entity
"I-CITY"      # Inside a city entity
"B-STATE"     # Beginning of a state entity
"I-STATE"     # Inside a state entity
"B-CITYSTATE" # Beginning of a city_state entity
"I-CITYSTATE" # Inside a city_state entity

Developed by: Mozilla
Language(s): English (en)
License: Apache-2.0
Fine-tuned from: distilbert/distilbert-base-uncased

Model Sources

Repository: Mozilla Smart Intent Project

Citation

If you use this model, please cite it as:

@misc{mozilla_distilbert_lora_ner,
  title       = {Fine-tuned LoRA Token Classifier on DistilBERT},
  author      = {Mozilla},
  year        = {2024},
  url         = {https://huggingface.co/Mozilla/distilbert-finetuned-LoRA-token-classifier},
  license     = {Apache-2.0}
}