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: 1.6254
- Bleu: 26.3346
- Gen Len: 26.6907
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 |
---|---|---|---|---|---|
2.3103 | 0.0355 | 10000 | 1.8454 | 22.058 | 27.7263 |
2.2141 | 0.0710 | 20000 | 1.7811 | 23.339 | 26.7147 |
2.176 | 0.1065 | 30000 | 1.7361 | 24.3234 | 27.125 |
2.139 | 0.1419 | 40000 | 1.7131 | 25.0888 | 26.8213 |
2.1084 | 0.1774 | 50000 | 1.6874 | 24.9992 | 26.824 |
2.0826 | 0.2129 | 60000 | 1.6685 | 25.7297 | 26.62 |
2.068 | 0.2484 | 70000 | 1.6485 | 25.9031 | 26.685 |
2.05 | 0.2839 | 80000 | 1.6371 | 26.143 | 26.8693 |
2.0331 | 0.3194 | 90000 | 1.6311 | 26.3038 | 26.5183 |
2.0273 | 0.3549 | 100000 | 1.6254 | 26.3346 | 26.6907 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.2+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1