small-prompt-compression

This model is a fine-tuned version of Falconsai/text_summarization on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1671
• Rouge1: 0.8011
• Rouge2: 0.6232
• Rougel: 0.7622
• Rougelsum: 0.7623
• Comp Ratio Mean: 0.8625
• Comp Ratio P90: 1.0
• Pct Violations: 0.0083

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3
• mixed_precision_training: Native AMP
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Comp Ratio Mean	Comp Ratio P90	Pct Violations
2.6954	1.0	4523	2.2681	0.7873	0.6009	0.7454	0.7455	0.8729	1.0	0.0183
2.3427	2.0	9046	2.1862	0.7959	0.6164	0.7562	0.7564	0.8672	1.0	0.0102
2.2822	3.0	13569	2.1671	0.8011	0.6232	0.7622	0.7623	0.8625	1.0	0.0083

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1