metadata
license: apache-2.0
base_model: google/long-t5-tglobal-xl
tags:
- generated_from_trainer
datasets:
- tau/scrolls
model-index:
- name: longt5_xl_sfd_4096_e10
results: []
longt5_xl_sfd_4096_e10
This model is a fine-tuned version of google/long-t5-tglobal-xl on the tau/scrolls summ_screen_fd dataset. It achieves the following results on the evaluation set:
- Loss: 2.3255
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: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.0332 | 0.97 | 14 | 2.5424 |
| 2.4105 | 1.95 | 28 | 2.3255 |
| 2.0496 | 2.99 | 43 | 2.3420 |
| 1.7473 | 3.97 | 57 | 2.3520 |
| 1.4007 | 4.94 | 71 | 2.4980 |
| 1.3809 | 5.98 | 86 | 2.4785 |
| 1.1153 | 6.96 | 100 | 2.7326 |
| 0.9129 | 8.0 | 115 | 2.9232 |
| 0.7118 | 8.97 | 129 | 3.0476 |
| 0.5883 | 9.74 | 140 | 3.3644 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3