mt5-small-multinews

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.6760
• Rouge1: 12.0734
• Rouge2: 4.3967
• Rougel: 10.3798
• Rougelsum: 11.183

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: 5.6e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
4.0379	1.0	1875	2.8647	11.7472	3.9041	10.0104	10.9935
3.1006	2.0	3750	2.7921	11.9174	4.1568	10.1817	11.1292
2.9625	3.0	5625	2.7340	11.8991	4.2439	10.2099	11.0833
2.8808	4.0	7500	2.7087	12.2156	4.3539	10.4789	11.3807
2.8298	5.0	9375	2.6980	12.0815	4.391	10.3708	11.2082
2.7949	6.0	11250	2.6671	12.1477	4.4187	10.4061	11.2805
2.7709	7.0	13125	2.6780	12.216	4.4787	10.4787	11.3018
2.7609	8.0	15000	2.6760	12.0734	4.3967	10.3798	11.183

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.1+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0