<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
synpre_set_1M_t5-base
This model is a fine-tuned version of t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0552
- Bleu: 99.1537
- Gen Len: 59.3765
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 128
- total_eval_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 10000
- training_steps: 80000
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
---|---|---|---|---|---|
9.3535 | 0.64 | 5000 | 9.1958 | 0.0166 | 87.7917 |
8.7053 | 1.28 | 10000 | 8.4713 | 0.1985 | 28.3437 |
8.2436 | 1.92 | 15000 | 7.8051 | 0.2801 | 29.6672 |
7.9472 | 2.56 | 20000 | 7.3145 | 0.5111 | 34.9599 |
3.6158 | 3.2 | 25000 | 2.1851 | 16.7092 | 41.2548 |
0.1081 | 3.84 | 30000 | 0.0552 | 99.1537 | 59.3765 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.13.0a0+d0d6b1f
- Datasets 2.12.0
- Tokenizers 0.13.3