Mitsuha21/t5-small-finetuned-xsum

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

• Train Loss: 1.8702
• Validation Loss: 1.4831
• Train Rouge1: 46.2436
• Train Rouge2: 26.6188
• Train Rougel: 42.7423
• Train Rougelsum: 42.6771
• Train Gen Len: 13.5220
• Epoch: 19

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:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Validation Loss	Train Rouge1	Train Rouge2	Train Rougel	Train Rougelsum	Train Gen Len	Epoch
4.6083	3.6627	20.0685	4.3594	18.1869	18.1861	18.0610	0
3.7845	3.2563	24.9949	6.4551	22.9249	22.9713	15.6122	1
3.4628	3.0291	28.1125	7.7006	25.8046	25.8162	14.4415	2
3.2508	2.8548	29.8432	8.8578	27.4137	27.4718	14.2268	3
3.0963	2.7116	31.7427	9.8558	28.6398	28.6626	14.3049	4
2.9616	2.5825	33.8958	11.4084	30.4746	30.5051	13.8512	5
2.8430	2.4661	34.9508	12.9830	31.7061	31.7622	13.6707	6
2.7381	2.3567	36.7926	14.5760	33.1279	33.1544	13.4049	7
2.6359	2.2565	36.4180	15.1004	33.2257	33.2507	13.8537	8
2.5543	2.1625	38.5778	17.1077	35.0027	35.0407	13.6	9
2.4596	2.0833	38.6852	17.2964	35.4659	35.4887	13.6902	10
2.3775	1.9966	40.1427	19.2079	36.3993	36.4188	13.4610	11
2.3072	1.9227	40.9101	19.7985	36.9402	36.9218	14.0049	12
2.2272	1.8442	42.0126	20.7988	37.9708	37.9803	13.8780	13
2.1612	1.7821	42.8467	22.2188	39.2833	39.2590	13.7195	14
2.1033	1.7130	44.1141	23.5104	40.0780	40.0617	14.1171	15
2.0401	1.6523	44.4919	24.3293	40.8943	40.8605	13.5756	16
1.9850	1.5952	44.9013	24.5225	41.0805	41.0159	13.3683	17
1.9253	1.5343	45.4848	25.9386	42.1817	42.1357	13.7902	18
1.8702	1.4831	46.2436	26.6188	42.7423	42.6771	13.5220	19

Framework versions

• Transformers 4.39.3
• TensorFlow 2.15.0
• Datasets 2.18.0
• Tokenizers 0.15.2