mt5-small-gigatrue-slovak-D-456
This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.6412
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: 0.0003
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.5083 | 0.1015 | 3000 | 2.8375 |
| 3.6752 | 0.2030 | 6000 | 2.7422 |
| 3.598 | 0.3044 | 9000 | 2.7021 |
| 3.5651 | 0.4059 | 12000 | 2.6863 |
| 3.5409 | 0.5074 | 15000 | 2.6730 |
| 3.5353 | 0.6089 | 18000 | 2.6633 |
| 3.5268 | 0.7104 | 21000 | 2.6614 |
| 3.5223 | 0.8119 | 24000 | 2.6543 |
| 3.5185 | 0.9133 | 27000 | 2.6546 |
| 3.515 | 1.0148 | 30000 | 2.6510 |
| 3.5138 | 1.1163 | 33000 | 2.6481 |
| 3.508 | 1.2178 | 36000 | 2.6456 |
| 3.5071 | 1.3193 | 39000 | 2.6427 |
| 3.5058 | 1.4207 | 42000 | 2.6433 |
| 3.5063 | 1.5222 | 45000 | 2.6441 |
| 3.5003 | 1.6237 | 48000 | 2.6423 |
| 3.5018 | 1.7252 | 51000 | 2.6439 |
| 3.4992 | 1.8267 | 54000 | 2.6423 |
| 3.5011 | 1.9282 | 57000 | 2.6425 |
| 3.5017 | 2.0296 | 60000 | 2.6415 |
| 3.5014 | 2.1311 | 63000 | 2.6422 |
| 3.4996 | 2.2326 | 66000 | 2.6411 |
| 3.4998 | 2.3341 | 69000 | 2.6423 |
| 3.5002 | 2.4356 | 72000 | 2.6412 |
| 3.5011 | 2.5370 | 75000 | 2.6415 |
| 3.5005 | 2.6385 | 78000 | 2.6411 |
| 3.4975 | 2.7400 | 81000 | 2.6413 |
| 3.5013 | 2.8415 | 84000 | 2.6411 |
| 3.5006 | 2.9430 | 87000 | 2.6412 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.5.1
- Datasets 3.2.0
- Tokenizers 0.20.3