metadata
library_name: transformers
license: mit
base_model: FacebookAI/xlm-roberta-large
tags:
- generated_from_trainer
model-index:
- name: relex_pre
results: []
relex_pre
This model is a fine-tuned version of FacebookAI/xlm-roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4173
- Macro F1: 0.9040
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 |
|---|---|---|---|---|
| 0.8776 | 1.0 | 1328 | 0.3809 | 0.7517 |
| 0.2962 | 2.0 | 2656 | 0.3053 | 0.8606 |
| 0.2057 | 3.0 | 3984 | 0.3026 | 0.8932 |
| 0.1408 | 4.0 | 5312 | 0.3286 | 0.9079 |
| 0.0961 | 5.0 | 6640 | 0.4013 | 0.8945 |
| 0.0628 | 6.0 | 7968 | 0.4145 | 0.9037 |
| 0.042 | 7.0 | 9296 | 0.4173 | 0.9040 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1