ViTay-sliding

This model is a fine-tuned version of FiveC/BartTay on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1400
• Sacrebleu: 20.7546

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: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Sacrebleu
0.1135	1.0	3951	0.1640	16.1738
0.0954	2.0	7902	0.1446	19.4432
0.0849	3.0	11853	0.1400	20.7546

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1