tankiit/credence-ablation-deberta-v3-n-heads-10-hatexplain
0
credence-ablation-deberta-v3-n-heads-10-hatexplain
CREDENCE checkpoint for hatexplain (toxicity) with backbone microsoft/deberta-v3-base (10 concept heads).
Paper: https://huggingface.co/papers/2604.24170 · arXiv: 2604.24170
Training run folder: head_ablation_deberta-v3_20251224_192526
Files
credence_checkpoint.pt— PyTorch checkpoint (model_state_dict,config,metadata, optimizer state)
Load with huggingface_hub
from huggingface_hub import hf_hub_download
import torch
path = hf_hub_download(repo_id="tankiit/credence-ablation-deberta-v3-n-heads-10-hatexplain", filename="credence_checkpoint.pt")
ckpt = torch.load(path, map_location="cpu", weights_only=False)
state = ckpt["model_state_dict"]
config = ckpt["config"]Load into your CREDENCE model implementation (see project credence.py).
Citation
@article{mukherjee2026credence,
title={Credal Concept Bottleneck Models for Epistemic--Aleatoric Uncertainty Decomposition},
author={Mukherjee, et al.},
journal={arXiv preprint arXiv:2604.24170},
year={2026}
}