xlm-roberta-base_lr5e-06_seed42_basic_eng_train

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0269
• Spearman Corr: 0.8037

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-06
• train_batch_size: 32
• eval_batch_size: 128
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Spearman Corr
0.118	2.33	200	0.0131	0.7608
0.0306	4.65	400	0.0163	0.7890
0.0248	6.98	600	0.0188	0.7988
0.0216	9.3	800	0.0272	0.7926
0.0201	11.63	1000	0.0228	0.8040
0.0171	13.95	1200	0.0208	0.8059
0.0158	16.28	1400	0.0271	0.8070
0.0145	18.6	1600	0.0236	0.8062
0.0139	20.93	1800	0.0279	0.8020
0.0132	23.26	2000	0.0243	0.8010
0.013	25.58	2200	0.0269	0.8037

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2