Instructions to use pangram/editlens_Llama-3.2-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use pangram/editlens_Llama-3.2-3B with PEFT:
from peft import PeftModel from transformers import AutoModelForSequenceClassification base_model = AutoModelForSequenceClassification.from_pretrained("meta-llama/Llama-3.2-3B") model = PeftModel.from_pretrained(base_model, "pangram/editlens_Llama-3.2-3B") - Notebooks
- Google Colab
- Kaggle
metadata
extra_gated_fields:
First Name: text
Last Name: text
Institution: text
Country: country
How do you intend to use this model?: text
I agree to use this model for non-commercial use ONLY: checkbox
base_model: meta-llama/Llama-3.2-3B
library_name: peft
tags:
- base_model:adapter:meta-llama/Llama-3.2-3B
- ai_detection
datasets:
- pangram/editlens_iclr
language:
- en
license: cc-by-nc-sa-4.0
Model Card for editlens_Llama-3.2-3B by Pangram
This model is a meta-llama/Llama-3.2-3B base model finetuned for AI detection according to the techniques described in the EditLens paper by Thai et al. (ICLR 2026)
Model Details
- Developed by: Pangram
- Language(s) (NLP): English
- License: CC BY-NC-SA 4.0
- Finetuned from model:
meta-llama/Llama-3.2-3B
Resources
- Repository: https://github.com/pangramlabs/EditLens
- Paper: https://arxiv.org/abs/2510.03154
Citation
BibTeX:
@misc{thai2025editlensquantifyingextentai,
title={EditLens: Quantifying the Extent of AI Editing in Text},
author={Katherine Thai and Bradley Emi and Elyas Masrour and Mohit Iyyer},
year={2025},
eprint={2510.03154},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2510.03154},
}