mt5-base-standard-baseline

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: 0.7040

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: 5e-05
• train_batch_size: 48
• eval_batch_size: 48
• seed: 42
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
4.2556	1.0	228	1.3857
1.3507	2.0	456	0.9755
1.1780	3.0	684	0.8687
1.0624	4.0	912	0.8247
1.0189	5.0	1140	0.7906
0.9578	6.0	1368	0.7692
0.9299	7.0	1596	0.7585
0.9043	8.0	1824	0.7489
0.8772	9.0	2052	0.7358
0.8519	10.0	2280	0.7282
0.8463	11.0	2508	0.7214
0.8330	12.0	2736	0.7177
0.8221	13.0	2964	0.7171
0.8045	14.0	3192	0.7134
0.8115	15.0	3420	0.7094
0.8016	16.0	3648	0.7062
0.7927	17.0	3876	0.7056
0.7913	18.0	4104	0.7051
0.7925	19.0	4332	0.7046
0.7881	20.0	4560	0.7040

Framework versions

• Transformers 5.3.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.3
• Tokenizers 0.22.2