mt5-small-gigatrue-layercut-D456
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: 2.6214
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 |
|---|---|---|---|
| 4.249 | 0.1015 | 3000 | 2.7814 |
| 3.4936 | 0.2030 | 6000 | 2.7083 |
| 3.4312 | 0.3044 | 9000 | 2.6841 |
| 3.4015 | 0.4059 | 12000 | 2.6656 |
| 3.3795 | 0.5074 | 15000 | 2.6510 |
| 3.3716 | 0.6089 | 18000 | 2.6407 |
| 3.3644 | 0.7104 | 21000 | 2.6435 |
| 3.3595 | 0.8119 | 24000 | 2.6354 |
| 3.3534 | 0.9133 | 27000 | 2.6312 |
| 3.3522 | 1.0148 | 30000 | 2.6333 |
| 3.3501 | 1.1163 | 33000 | 2.6283 |
| 3.3451 | 1.2178 | 36000 | 2.6231 |
| 3.344 | 1.3193 | 39000 | 2.6240 |
| 3.3423 | 1.4207 | 42000 | 2.6217 |
| 3.3419 | 1.5222 | 45000 | 2.6252 |
| 3.3366 | 1.6237 | 48000 | 2.6247 |
| 3.3399 | 1.7252 | 51000 | 2.6215 |
| 3.3344 | 1.8267 | 54000 | 2.6219 |
| 3.337 | 1.9282 | 57000 | 2.6212 |
| 3.3394 | 2.0296 | 60000 | 2.6211 |
| 3.3393 | 2.1311 | 63000 | 2.6212 |
| 3.3353 | 2.2326 | 66000 | 2.6227 |
| 3.3369 | 2.3341 | 69000 | 2.6227 |
| 3.3348 | 2.4356 | 72000 | 2.6220 |
| 3.3358 | 2.5370 | 75000 | 2.6211 |
| 3.3377 | 2.6385 | 78000 | 2.6211 |
| 3.3336 | 2.7400 | 81000 | 2.6218 |
| 3.3359 | 2.8415 | 84000 | 2.6216 |
| 3.3368 | 2.9430 | 87000 | 2.6214 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.5.1
- Datasets 3.2.0
- Tokenizers 0.20.3