mt5-base-standard-baseline-final

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: 0.7195

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• 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
23.4668	1.0	127	3.0561
3.7655	2.0	254	1.1775
2.8581	3.0	381	0.9833
2.4228	4.0	508	0.9030
2.2903	5.0	635	0.8595
2.1044	6.0	762	0.8174
2.0442	7.0	889	0.8005
1.9340	8.0	1016	0.7781
1.8886	9.0	1143	0.7653
1.8404	10.0	1270	0.7559
1.8040	11.0	1397	0.7488
1.7736	12.0	1524	0.7407
1.7289	13.0	1651	0.7358
1.7204	14.0	1778	0.7309
1.7002	15.0	1905	0.7256
1.6891	16.0	2032	0.7269
1.6609	17.0	2159	0.7202
1.6685	18.0	2286	0.7199
1.6618	19.0	2413	0.7197
1.6518	20.0	2540	0.7195

Framework versions

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