metadata
license: mit
library_name: peft
tags:
- generated_from_trainer
base_model: facebook/bart-large-cnn
datasets:
- samsum
model-index:
- name: bart-cnn-samsum-peft
results: []
bart-cnn-samsum-peft
This model is a fine-tuned version of facebook/bart-large-cnn on the samsum dataset. It achieves the following results on the evaluation set:
- Loss: 0.5166
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.9416 | 1.0 | 37 | 0.7712 |
| 0.6612 | 2.0 | 74 | 0.6450 |
| 0.5657 | 3.0 | 111 | 0.5918 |
| 0.5758 | 4.0 | 148 | 0.5598 |
| 0.499 | 5.0 | 185 | 0.5426 |
| 0.4521 | 6.0 | 222 | 0.5314 |
| 0.4618 | 7.0 | 259 | 0.5244 |
| 0.4838 | 8.0 | 296 | 0.5202 |
| 0.4983 | 9.0 | 333 | 0.5175 |
| 0.4688 | 10.0 | 370 | 0.5166 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1