xlm-roberta-large-sentence-splitter

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

• Loss: 0.0102
• 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	0.0163	0.8264
No log	2.0	98	0.0189	0.8159
No log	3.0	147	0.0091	0.8791
No log	4.0	196	0.0154	0.8908
No log	5.0	245	0.0134	0.9784
No log	6.0	294	0.0163	0.8466
No log	7.0	343	0.0019	0.9922
No log	8.0	392	0.0028	0.9846
No log	9.0	441	0.0139	0.9922
No log	10.0	490	0.0016	0.9938
0.0207	11.0	539	0.0076	0.9922
0.0207	12.0	588	0.0087	0.9953
0.0207	13.0	637	0.0095	0.9922
0.0207	14.0	686	0.0086	0.9938
0.0207	15.0	735	0.0083	0.9938
0.0207	16.0	784	0.0020	0.9922
0.0207	17.0	833	0.0023	0.9938
0.0207	18.0	882	0.0029	0.9922
0.0207	19.0	931	0.0085	0.9938
0.0207	20.0	980	0.0130	0.9907
0.0007	21.0	1029	0.0126	0.9907
0.0007	22.0	1078	0.0106	0.9922
0.0007	23.0	1127	0.0103	0.9922
0.0007	24.0	1176	0.0103	0.9922
0.0007	25.0	1225	0.0102	0.9922
0.0007	26.0	1274	0.0102	0.9922
0.0007	27.0	1323	0.0102	0.9922
0.0007	28.0	1372	0.0102	0.9922
0.0007	29.0	1421	0.0102	0.9922
0.0007	30.0	1470	0.0102	0.9922

Framework versions

• Transformers 4.55.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4