model

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: 3.2858

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: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.2251	0.0533	1000	4.1491
2.7413	0.1067	2000	3.7912
2.5416	0.16	3000	3.6801
2.371	0.2133	4000	3.6439
2.2968	0.2667	5000	3.5301
2.1989	0.32	6000	3.3905
2.0841	0.3733	7000	3.5244
2.0032	0.4267	8000	3.3268
1.9618	0.48	9000	3.3207
1.9114	0.5333	10000	3.4544
1.8472	0.5867	11000	3.2520
1.8068	0.64	12000	3.3389
1.7692	0.6933	13000	3.2428
1.7236	0.7467	14000	3.3926
1.7219	0.8	15000	3.2721
1.6838	0.8533	16000	3.2671
1.6771	0.9067	17000	3.2732
1.6531	0.96	18000	3.2858

Framework versions

• Transformers 4.48.1
• Pytorch 2.1.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0