Edit model card

gbert-base-finetuned-twitter_

This model is a fine-tuned version of deepset/gbert-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6651

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: 192
  • eval_batch_size: 192
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss
2.1933 1.0 4180 1.9612
2.0051 2.0 8360 1.8795
1.939 3.0 12540 1.8310
1.8928 4.0 16720 1.8013
1.8594 5.0 20900 1.7730
1.8336 6.0 25080 1.7702
1.8145 7.0 29260 1.7449
1.7963 8.0 33440 1.7277
1.7806 9.0 37620 1.7105
1.7682 10.0 41800 1.7061
1.7584 11.0 45980 1.7041
1.7454 12.0 50160 1.6899
1.7374 13.0 54340 1.6850
1.7295 14.0 58520 1.6856
1.7232 15.0 62700 1.6819
1.715 16.0 66880 1.6730
1.7101 17.0 71060 1.6723
1.7057 18.0 75240 1.6655
1.7038 19.0 79420 1.6617
1.702 20.0 83600 1.6625

Framework versions

  • Transformers 4.32.1
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.4
  • Tokenizers 0.13.3
Downloads last month
6

Finetuned from