mt5-base-bias-aware-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: 0.6808

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
2.0090	1.0	889	0.7767
1.6572	2.0	1778	0.7032
1.4638	3.0	2667	0.6705
1.3440	4.0	3556	0.6528
1.2512	5.0	4445	0.6481
1.1572	6.0	5334	0.6422
1.0987	7.0	6223	0.6384
1.0329	8.0	7112	0.6442
0.9974	9.0	8001	0.6441
0.9520	10.0	8890	0.6489
0.9086	11.0	9779	0.6554
0.8743	12.0	10668	0.6613
0.8472	13.0	11557	0.6635
0.8171	14.0	12446	0.6664
0.8001	15.0	13335	0.6704
0.7833	16.0	14224	0.6745
0.7651	17.0	15113	0.6777
0.7519	18.0	16002	0.6797
0.7533	19.0	16891	0.6794
0.7485	20.0	17780	0.6808

Framework versions

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