mt5-base_V25775_V44105

This model is a fine-tuned version of emilstabil/mt5-base_V25775 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1019
• Rouge1: 30.2631
• Rouge2: 10.8564
• Rougel: 20.9297
• Rougelsum: 24.8312
• Gen Len: 80.9356

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: 3
• eval_batch_size: 3
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.32	0.81	500	2.1435	28.7386	10.9481	20.3975	23.8195	74.2704
2.2393	1.61	1000	2.1053	29.1856	10.7042	20.4864	24.1221	75.515
2.2124	2.42	1500	2.1157	28.6845	10.9397	20.4075	23.9154	74.8627
2.1635	3.23	2000	2.1232	28.8373	10.8364	20.4743	24.0269	74.1845
2.1148	4.03	2500	2.1149	29.0484	11.0898	20.6711	24.0963	73.9571
2.0904	4.84	3000	2.1101	29.5911	11.2027	20.883	24.3776	76.8412
2.0598	5.65	3500	2.1212	29.5276	10.8551	20.5466	24.1469	78.4506
2.0596	6.45	4000	2.1368	29.8832	10.9578	20.7962	24.4686	80.3777
2.0135	7.26	4500	2.1173	29.5314	10.6881	20.375	24.2483	81.5751
2.0085	8.06	5000	2.1050	29.7932	11.0481	20.8481	24.5598	78.5708
2.0006	8.87	5500	2.1233	30.4225	11.3125	21.1509	24.9171	81.3648
1.9888	9.68	6000	2.1067	29.9013	10.7672	20.6523	24.5878	78.7897
1.9496	10.48	6500	2.1036	29.7453	10.9583	20.7396	24.3824	78.7425
1.9513	11.29	7000	2.1125	29.5484	10.752	20.4861	24.3097	79.0558
1.9476	12.1	7500	2.1014	29.6296	10.8252	20.6412	24.2908	76.1202
1.9294	12.9	8000	2.1102	29.9456	10.9121	20.8077	24.5787	79.515
1.9036	13.71	8500	2.0977	30.1173	10.9352	20.9176	24.9725	80.9056
1.9415	14.52	9000	2.1011	29.9247	10.8223	20.7609	24.6858	81.103
1.8959	15.32	9500	2.0998	29.8002	10.6206	20.5674	24.6966	80.4549
1.9356	16.13	10000	2.1038	30.355	11.0359	21.0347	25.0475	80.8927
1.8958	16.94	10500	2.1029	30.3957	11.0562	21.1067	25.1431	82.1588
1.9093	17.74	11000	2.1002	30.4669	10.9894	20.9725	24.9598	81.1888
1.8969	18.55	11500	2.1045	30.4956	10.9426	20.9578	24.9973	81.824
1.8971	19.35	12000	2.1019	30.2631	10.8564	20.9297	24.8312	80.9356

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0
• Datasets 2.12.0
• Tokenizers 0.13.3