xlm-roberta-base_lr0.0001_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: 0.0235
• Spearman Corr: nan

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: 0.0001
• 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
0.0684	2.33	200	0.0235	-0.1233
0.0564	4.65	400	0.0243	nan
0.0545	6.98	600	0.0271	0.1235
0.0539	9.3	800	0.0234	nan
0.0532	11.63	1000	0.0243	-0.0724
0.0533	13.95	1200	0.0233	0.0967
0.0519	16.28	1400	0.0244	0.0055
0.0519	18.6	1600	0.0243	0.0838
0.0515	20.93	1800	0.0233	nan
0.0516	23.26	2000	0.0253	nan
0.0511	25.58	2200	0.0245	-0.0083
0.0503	27.91	2400	0.0235	nan

Framework versions

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