<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert-base-uncased-finetuned-discord
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.6981
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.4692 | 1.0 | 329 | 3.0852 |
3.1111 | 2.0 | 658 | 2.9595 |
2.9821 | 3.0 | 987 | 2.8469 |
2.9075 | 4.0 | 1316 | 2.8171 |
2.8413 | 5.0 | 1645 | 2.7734 |
2.8313 | 6.0 | 1974 | 2.7486 |
2.7857 | 7.0 | 2303 | 2.7608 |
2.7779 | 8.0 | 2632 | 2.7335 |
2.752 | 9.0 | 2961 | 2.7069 |
2.7529 | 10.0 | 3290 | 2.7005 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3