fassabilf's picture
Model save
d8cf373 verified
metadata
library_name: transformers
license: apache-2.0
base_model: cross-encoder/nli-deberta-v3-large
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: enli-deberta-v3-large_10
    results: []

enli-deberta-v3-large_10

This model is a fine-tuned version of cross-encoder/nli-deberta-v3-large on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8908
  • Accuracy: 0.9083
  • F1: 0.9088

Model description

More information needed

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: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy F1
0.2817 1.0 3516 0.2948 0.9042 0.9045
0.1676 2.0 7032 0.3862 0.9029 0.9033
0.1199 3.0 10548 0.5174 0.9017 0.9021
0.0501 4.0 14064 0.6014 0.9039 0.9043
0.0489 5.0 17580 0.7007 0.9034 0.9039
0.0173 6.0 21096 0.7448 0.9049 0.9056
0.0142 7.0 24612 0.7086 0.9062 0.9068
0.0003 8.0 28128 0.8314 0.9074 0.9079
0.0059 9.0 31644 0.8942 0.9078 0.9083
0.0 10.0 35160 0.8908 0.9083 0.9088

Framework versions

  • Transformers 4.52.4
  • Pytorch 2.7.1+cu128
  • Datasets 3.6.0
  • Tokenizers 0.21.1