tiny-mlm-glue-qnli-target-glue-cola

This model is a fine-tuned version of muhtasham/tiny-mlm-glue-qnli on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7322
• Matthews Correlation: 0.1353

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: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss	Matthews Correlation
0.6099	1.87	500	0.6209	0.0
0.6009	3.73	1000	0.6169	0.0
0.5819	5.6	1500	0.6196	0.0545
0.5519	7.46	2000	0.6391	0.0997
0.5226	9.33	2500	0.6657	0.1182
0.5061	11.19	3000	0.6671	0.1357
0.4831	13.06	3500	0.6787	0.1205
0.4652	14.93	4000	0.7167	0.1264
0.4443	16.79	4500	0.7322	0.1353

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu116
• Datasets 2.8.1.dev0
• Tokenizers 0.13.2