samsum_model_t5_small_10_epochs

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

• Loss: 1.8701
• Rouge1: 0.4055
• Rouge2: 0.1762
• Rougel: 0.3372
• Rougelsum: 0.337
• Gen Len: 16.4738

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 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
No log	1.0	200	1.9528	0.3844	0.1567	0.3184	0.3182	16.1362
No log	2.0	400	1.9221	0.3885	0.1613	0.3212	0.321	16.3325
2.0996	3.0	600	1.9072	0.3936	0.1661	0.3264	0.3259	16.2288
2.0996	4.0	800	1.8930	0.3984	0.1678	0.3295	0.3292	16.3375
2.0297	5.0	1000	1.8860	0.4005	0.1708	0.333	0.3329	16.355
2.0297	6.0	1200	1.8780	0.4023	0.1726	0.3341	0.3342	16.3375
2.0297	7.0	1400	1.8738	0.4025	0.1723	0.3347	0.3346	16.4275
1.9894	8.0	1600	1.8701	0.4064	0.1757	0.3369	0.3369	16.495
1.9894	9.0	1800	1.8706	0.4061	0.1767	0.3375	0.3375	16.4825
1.9735	10.0	2000	1.8701	0.4055	0.1762	0.3372	0.337	16.4738

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3