xlm-roberta-base_lr2e-05_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: 0.0266
• Spearman Corr: 0.8166

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: 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
0.0451	2.33	200	0.0186	0.7909
0.0202	4.65	400	0.0282	0.8043
0.0153	6.98	600	0.0223	0.8111
0.0117	9.3	800	0.0272	0.8021
0.0096	11.63	1000	0.0267	0.8105
0.007	13.95	1200	0.0253	0.8226
0.0061	16.28	1400	0.0269	0.8107
0.0052	18.6	1600	0.0252	0.8172
0.0047	20.93	1800	0.0264	0.8179
0.0044	23.26	2000	0.0247	0.8206
0.004	25.58	2200	0.0266	0.8166

Framework versions

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