xlm-roberta-base-sentence-splitter

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0078
• F1: 0.9922

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: 8
• eval_batch_size: 8
• 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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	49	0.0028	0.9831
No log	2.0	98	0.0021	0.9877
No log	3.0	147	0.0046	0.9907
No log	4.0	196	0.0014	0.9953
No log	5.0	245	0.0024	0.9892
No log	6.0	294	0.0064	0.9891
No log	7.0	343	0.0015	0.9938
No log	8.0	392	0.0030	0.9922
No log	9.0	441	0.0018	0.9922
No log	10.0	490	0.0106	0.9891
0.0148	11.0	539	0.0030	0.9892
0.0148	12.0	588	0.0026	0.9907
0.0148	13.0	637	0.0024	0.9907
0.0148	14.0	686	0.0022	0.9892
0.0148	15.0	735	0.0085	0.9907
0.0148	16.0	784	0.0019	0.9907
0.0148	17.0	833	0.0021	0.9922
0.0148	18.0	882	0.0106	0.9891
0.0148	19.0	931	0.0118	0.9891
0.0148	20.0	980	0.0047	0.9876
0.0003	21.0	1029	0.0073	0.9891
0.0003	22.0	1078	0.0088	0.9891
0.0003	23.0	1127	0.0152	0.9907
0.0003	24.0	1176	0.0057	0.9922
0.0003	25.0	1225	0.0043	0.9938
0.0003	26.0	1274	0.0047	0.9922
0.0003	27.0	1323	0.0084	0.9953
0.0003	28.0	1372	0.0080	0.9922
0.0003	29.0	1421	0.0078	0.9922
0.0003	30.0	1470	0.0078	0.9922

Framework versions

• Transformers 4.55.2
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4