t5-small-finetuned-xsum-custom

This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.5478
• Rouge1: 28.4804
• Rouge2: 7.7367
• Rougel: 22.7607
• Rougelsum: 22.762

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: 4e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use OptimizerNames.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: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
2.9745	1.0	999	2.6463	25.8883	6.5128	20.4979	20.4769
2.7924	2.0	1998	2.5992	27.3518	7.3916	21.7638	21.7476
2.7061	3.0	2997	2.5763	27.7159	7.5086	22.188	22.1916
2.6502	4.0	3996	2.5637	28.175	7.7661	22.6274	22.6179
2.6044	5.0	4995	2.5571	28.2348	7.7937	22.6196	22.6568
2.5781	6.0	5994	2.5526	28.319	7.7453	22.6005	22.6044
2.5618	7.0	6993	2.5488	28.4962	7.7803	22.7827	22.803
2.5441	8.0	7992	2.5478	28.4804	7.7367	22.7607	22.762

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1