DeBERTa-v3-large fine-tuned on MNLI
This model is a fine-tuned version of microsoft/deberta-v3-large on the GLUE MNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.6763
- Accuracy: 0.8949
Model description
DeBERTa improves the BERT and RoBERTa models using disentangled attention and enhanced mask decoder. With those two improvements, DeBERTa out perform RoBERTa on a majority of NLU tasks with 80GB training data.
In DeBERTa V3, we further improved the efficiency of DeBERTa using ELECTRA-Style pre-training with Gradient Disentangled Embedding Sharing. Compared to DeBERTa, our V3 version significantly improves the model performance on downstream tasks. You can find more technique details about the new model from our paper.
Please check the official repository for more implementation details and updates.
The DeBERTa V3 large model comes with 24 layers and a hidden size of 1024. It has 304M backbone parameters with a vocabulary containing 128K tokens which introduces 131M parameters in the Embedding layer. This model was trained using the 160GB data as DeBERTa V2.
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.3676 | 1.0 | 24544 | 0.3761 | 0.8681 |
0.2782 | 2.0 | 49088 | 0.3605 | 0.8881 |
0.1986 | 3.0 | 73632 | 0.4672 | 0.8894 |
0.1299 | 4.0 | 98176 | 0.5248 | 0.8967 |
0.0643 | 5.0 | 122720 | 0.6489 | 0.8999 |
Framework versions
- Transformers 4.13.0.dev0
- Pytorch 1.10.0+cu111
- Datasets 1.16.1
- Tokenizers 0.10.3
- Downloads last month
- 22
Dataset used to train mrm8488/deberta-v3-large-finetuned-mnli
Evaluation results
- Accuracy on GLUE MNLIself-reported0.895
- Accuracy on glueverified0.900
- Precision Macro on glueverified0.900
- Precision Micro on glueverified0.900
- Precision Weighted on glueverified0.901
- Recall Macro on glueverified0.900
- Recall Micro on glueverified0.900
- Recall Weighted on glueverified0.900
- F1 Macro on glueverified0.900
- F1 Micro on glueverified0.900