mt5-base-translation-spa-pbb

This model is a fine-tuned version of google/mt5-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.0646
• Bleu: 0.5194
• Gen Len: 5.3808

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
9.0597	1.0	304	7.8135	0.0148	4.6469
6.2294	2.0	608	4.9617	0.0	3.9209
4.8326	3.0	912	4.0494	0.0	4.3808
4.582	4.0	1216	3.7069	0.0	5.3979
5.4762	5.0	1520	3.5463	0.0	5.6759
4.3875	6.0	1824	3.4731	0.0	5.6258
4.2873	7.0	2128	3.3832	0.0	5.5455
4.1326	8.0	2432	3.3424	0.0	5.4756
3.5728	9.0	2736	3.2956	0.0	5.1792
3.1873	10.0	3040	3.2690	0.0	5.5903
2.9436	11.0	3344	3.2465	0.1237	5.7655
4.2955	12.0	3648	3.2054	0.1741	5.4466
3.8722	13.0	3952	3.1764	0.1887	5.2161
3.5391	14.0	4256	3.1688	0.0951	5.7312
3.8012	15.0	4560	3.1480	0.1948	5.2964
3.1148	16.0	4864	3.1401	0.2397	5.7589
3.2699	17.0	5168	3.1186	0.33	5.386
4.3355	18.0	5472	3.1092	0.4637	5.1383
3.5792	19.0	5776	3.0966	0.3286	5.4374
3.1429	20.0	6080	3.0923	0.418	5.2964
3.4155	21.0	6384	3.0900	0.3938	5.4848
3.4515	22.0	6688	3.0755	0.4062	5.4124
2.8244	23.0	6992	3.0717	0.4218	5.3663
2.9253	24.0	7296	3.0663	0.3633	5.5692
2.1757	25.0	7600	3.0640	0.4768	5.4282
2.9356	26.0	7904	3.0646	0.5194	5.3808

Framework versions

• Transformers 4.35.2
• Pytorch 1.13.1+cu117
• Datasets 2.15.0
• Tokenizers 0.15.0