tiny-mlm-glue-qnli-from-scratch-custom-tokenizer-target-glue-cola

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

• Loss: 0.8016
• Matthews Correlation: 0.0716

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
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Matthews Correlation
0.6104	1.87	500	0.6207	0.0
0.6069	3.73	1000	0.6185	0.0
0.5994	5.6	1500	0.6141	-0.0207
0.5555	7.46	2000	0.6537	0.0485
0.5134	9.33	2500	0.6886	0.0900
0.4922	11.19	3000	0.6950	0.0998
0.4643	13.06	3500	0.7414	0.1108
0.4497	14.93	4000	0.7671	0.0889
0.4315	16.79	4500	0.7826	0.0760
0.421	18.66	5000	0.8016	0.0716

Framework versions

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