metadata
license: apache-2.0
base_model: google-t5/t5-small
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: ft-wmt14
results: []
ft-wmt14
This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.7469
- Bleu: 23.6596
- Gen Len: 27.526
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: 0.0005
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 100000
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.7738 | 0.2778 | 10000 | 1.9146 | 20.1598 | 28.1563 |
| 1.6498 | 0.5556 | 20000 | 1.8550 | 21.4167 | 27.853 |
| 1.5903 | 0.8333 | 30000 | 1.8277 | 22.604 | 27.7613 |
| 1.5151 | 1.1111 | 40000 | 1.8128 | 22.1273 | 27.3187 |
| 1.4866 | 1.3889 | 50000 | 1.7999 | 22.8295 | 27.419 |
| 1.4696 | 1.6667 | 60000 | 1.7810 | 22.9923 | 27.7387 |
| 1.4508 | 1.9444 | 70000 | 1.7654 | 23.1046 | 27.7057 |
| 1.4053 | 2.2222 | 80000 | 1.7587 | 23.5079 | 27.643 |
| 1.3956 | 2.5 | 90000 | 1.7525 | 23.3848 | 27.6637 |
| 1.3903 | 2.7778 | 100000 | 1.7469 | 23.6596 | 27.526 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1