# MNLI Fine-tuning

DeBERTa V2 XXLarge is an enhanced BERT variant based on the disentangled attention mechanism, surpassing RoBERTa and XLNet in natural language understanding tasks with 1.5 billion parameters

This model is a PyTorch pretrained model obtained by converting TensorFlow checkpoints from the official Google BERT repository, originating from the paper 'Well-Read Students Learn Better: On the Importance of Pre-training Compact Models,' and trained on the MNLI dataset.

An enhanced BERT decoding model based on the decoupled attention mechanism, a large-scale version fine-tuned on the MNLI task.