bge-reranker-v2-m3-finetuned

This model is a fine-tuned version of BAAI/bge-reranker-v2-m3 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2950
• Pointwise Accuracy: 0.919
• Pointwise F1: 0.7990
• Pointwise Auc: 0.9567
• Group Top1 Accuracy: 0.885

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: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• 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
• lr_scheduler_warmup_steps: 0.1
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Pointwise Accuracy	Pointwise F1	Pointwise Auc	Group Top1 Accuracy
1.8756	1.0	619	0.2950	0.919	0.7990	0.9567	0.885

Framework versions

• Transformers 5.13.0
• Pytorch 2.10.0+cu128
• Datasets 5.0.0
• Tokenizers 0.22.2