roberta-large-unlabeled-labeled-gab-reddit-task-semeval2023-t10-210000sample
This model is a fine-tuned version of HPL/roberta-large-unlabeled-labeled-gab-reddit-task-semeval2023-t10-150000sample on the None dataset.
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: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
Framework versions
- Transformers 4.13.0
- Pytorch 1.13.0+cu116
- Datasets 2.8.0
- Tokenizers 0.10.3