distilbert_add_GLUE_Experiment_stsb_256

This model is a fine-tuned version of distilbert-base-uncased on the GLUE STSB dataset. It achieves the following results on the evaluation set:

• Loss: 2.2898
• Pearson: 0.0723
• Spearmanr: 0.0744
• Combined Score: 0.0733

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: 256
• eval_batch_size: 256
• seed: 10
• distributed_type: multi-GPU
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Pearson	Spearmanr	Combined Score
5.798	1.0	23	3.1859	nan	nan	nan
3.2592	2.0	46	2.3672	nan	nan	nan
2.3588	3.0	69	2.3366	nan	nan	nan
2.1815	4.0	92	2.3354	nan	nan	nan
2.1676	5.0	115	2.3685	0.0701	0.0628	0.0665
2.1604	6.0	138	2.3425	0.0799	0.0728	0.0764
2.1203	7.0	161	2.2898	0.0723	0.0744	0.0733
1.8844	8.0	184	2.7739	0.0606	0.0839	0.0723
1.7797	9.0	207	2.6237	0.0817	0.0949	0.0883
1.7003	10.0	230	2.7269	0.0957	0.1082	0.1020
1.5943	11.0	253	2.6580	0.1212	0.1276	0.1244
1.5603	12.0	276	2.5384	0.1412	0.1422	0.1417

Framework versions

• Transformers 4.26.0
• Pytorch 1.14.0a0+410ce96
• Datasets 2.8.0
• Tokenizers 0.13.2