metadata
license: apache-2.0
base_model: facebook/bart-large
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: emnlp_test_9clusters_msmarco_bart_large
results: []
emnlp_test_9clusters_msmarco_bart_large
This model is a fine-tuned version of facebook/bart-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8042
- Bleu: 70.172
- Gen Len: 10.3243
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: 2e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 2.96 | 1.0 | 515 | 1.7686 | 33.692 | 11.1826 |
| 1.5254 | 2.0 | 1030 | 1.2312 | 54.1545 | 10.2574 |
| 1.0521 | 3.0 | 1545 | 1.0201 | 62.9221 | 10.3148 |
| 0.8446 | 4.0 | 2060 | 0.8635 | 67.2188 | 10.3663 |
| 0.7339 | 5.0 | 2575 | 0.8200 | 69.3947 | 10.354 |
| 0.6745 | 6.0 | 3090 | 0.8042 | 70.172 | 10.3243 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1