crossencoder-km1
This model is a fine-tuned version of cross-encoder/stsb-TinyBERT-L-4 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0110
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: 5e-05
- train_batch_size: 100
- eval_batch_size: 80
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.2478 | 1.0 | 20 | 6.6948 |
| 3.8026 | 2.0 | 40 | 2.8643 |
| 0.9993 | 3.0 | 60 | 0.8714 |
| 0.2986 | 4.0 | 80 | 0.2379 |
| 0.1161 | 5.0 | 100 | 0.0786 |
| 0.0414 | 6.0 | 120 | 0.0461 |
| 0.0218 | 7.0 | 140 | 0.0250 |
| 0.0144 | 8.0 | 160 | 0.0140 |
| 0.0101 | 9.0 | 180 | 0.0122 |
| 0.0083 | 10.0 | 200 | 0.0120 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.0.1
- Datasets 2.17.0
- Tokenizers 0.15.2