metadata
license: mit
base_model: deepset/gbert-base
tags:
- generated_from_trainer
model-index:
- name: gbert-base-finetuned-twitter
results: []
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: 1.7380
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.194 | 1.0 | 4180 | 1.9622 |
| 2.0075 | 2.0 | 8360 | 1.8813 |
| 1.9429 | 3.0 | 12540 | 1.8339 |
| 1.8985 | 4.0 | 16720 | 1.8057 |
| 1.8676 | 5.0 | 20900 | 1.7801 |
| 1.8446 | 6.0 | 25080 | 1.7793 |
| 1.829 | 7.0 | 29260 | 1.7580 |
| 1.815 | 8.0 | 33440 | 1.7445 |
| 1.8048 | 9.0 | 37620 | 1.7319 |
| 1.7997 | 10.0 | 41800 | 1.7331 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3