pegasus-large
This model is a fine-tuned version of google/pegasus-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6537
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: 3e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: polynomial
- lr_scheduler_warmup_steps: 500
- training_steps: 20000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
4.7374 | 0.0 | 500 | 3.7869 |
0.6947 | 0.01 | 1000 | 0.7657 |
0.7536 | 0.01 | 1500 | 0.7410 |
0.6689 | 0.01 | 2000 | 0.7213 |
0.662 | 0.02 | 2500 | 0.7120 |
0.7537 | 0.02 | 3000 | 0.7015 |
0.79 | 0.02 | 3500 | 0.6990 |
0.6765 | 0.03 | 4000 | 0.6920 |
0.7353 | 0.03 | 4500 | 0.6881 |
0.633 | 0.03 | 5000 | 0.6858 |
0.5015 | 0.04 | 5500 | 0.6828 |
0.8227 | 0.04 | 6000 | 0.6806 |
0.832 | 0.05 | 6500 | 0.6790 |
0.587 | 0.05 | 7000 | 0.6743 |
0.6121 | 0.05 | 7500 | 0.6726 |
0.621 | 0.06 | 8000 | 0.6713 |
0.5112 | 0.06 | 8500 | 0.6714 |
0.6596 | 0.06 | 9000 | 0.6677 |
0.7421 | 0.07 | 9500 | 0.6682 |
0.5891 | 0.07 | 10000 | 0.6655 |
0.596 | 0.07 | 10500 | 0.6660 |
0.7527 | 0.08 | 11000 | 0.6639 |
0.8404 | 0.08 | 11500 | 0.6620 |
0.6896 | 0.08 | 12000 | 0.6624 |
0.7312 | 0.09 | 12500 | 0.6598 |
0.7061 | 0.09 | 13000 | 0.6591 |
0.5983 | 0.09 | 13500 | 0.6594 |
0.659 | 0.1 | 14000 | 0.6587 |
0.8656 | 0.1 | 14500 | 0.6568 |
0.7991 | 0.1 | 15000 | 0.6571 |
0.6637 | 0.11 | 15500 | 0.6571 |
0.5115 | 0.11 | 16000 | 0.6558 |
0.6464 | 0.11 | 16500 | 0.6566 |
0.6673 | 0.12 | 17000 | 0.6550 |
0.6477 | 0.12 | 17500 | 0.6544 |
0.8145 | 0.13 | 18000 | 0.6542 |
0.6216 | 0.13 | 18500 | 0.6537 |
0.943 | 0.13 | 19000 | 0.6539 |
0.788 | 0.14 | 19500 | 0.6538 |
0.5921 | 0.14 | 20000 | 0.6537 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.15.1
- Downloads last month
- 10
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for alshelt/ctrlsum-pegasus-large
Base model
google/pegasus-large