xnli_m_bert_only_bg

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

• Loss: 1.2991
• Accuracy: 0.7365

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: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6597	1.0	3068	0.6952	0.7052
0.5758	2.0	6136	0.6158	0.7422
0.4912	3.0	9204	0.6293	0.7486
0.4073	4.0	12272	0.6818	0.7353
0.3286	5.0	15340	0.7461	0.7438
0.2562	6.0	18408	0.8900	0.7337
0.1959	7.0	21476	0.9912	0.7333
0.1483	8.0	24544	1.0983	0.7285
0.1097	9.0	27612	1.1904	0.7333
0.0811	10.0	30680	1.2991	0.7365

Framework versions

• Transformers 4.24.0
• Pytorch 1.13.0
• Datasets 2.6.1
• Tokenizers 0.13.1