bge-reranker-v2-m3-finetune

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: 0.3696
• Spearman: nan

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Spearman
No log	0	0	1.0280	0.0247
0.4622	0.1639	1000	0.3804	0.1140
0.4779	0.3278	2000	0.3672	0.0431
0.3758	0.4916	3000	0.3703	nan
0.1635	0.6555	4000	0.3711	nan
0.3221	0.8194	5000	0.3684	nan
0.3741	0.9833	6000	0.3674	-0.0151
0.502	1.1472	7000	0.3676	nan
0.6405	1.3110	8000	0.3675	nan
0.6499	1.4749	9000	0.3683	nan
0.3073	1.6388	10000	0.3717	-0.0190
0.2343	1.8027	11000	0.3700	-0.0153
0.6064	1.9666	12000	0.3729	nan
0.2455	2.1304	13000	0.3700	-0.0119
0.2484	2.2943	14000	0.3690	nan
0.3715	2.4582	15000	0.3695	-0.0078
0.2333	2.6221	16000	0.3696	nan
0.4614	2.7860	17000	0.3696	nan
0.2435	2.9499	18000	0.3696	nan

Framework versions

• Transformers 4.56.2
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.22.1