bert-base-multilingual-cased-finetuned-PQuAD-3epochs

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

• Loss: 0.8517
• {'exact': 73.3816545863534, 'f1': 86.40922381247157, 'total': 8002, 'HasAns_exact': 67.41130091984232, 'HasAns_f1': 84.53459411093914, 'HasAns_total': 6088, 'NoAns_exact': 92.37199582027168, 'NoAns_f1': 92.37199582027168, 'NoAns_total': 1914, 'best_exact': 73.3816545863534, 'best_exact_thresh': 0.0, 'best_f1': 86.40922381247229, 'best_f1_thresh': 0.0}

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.7978	1.0	4051	0.8395
0.5898	2.0	8102	0.8184
0.4274	3.0	12153	0.8517

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1