add_BERT_24_qnli

This model is a fine-tuned version of gokuls/add_bert_12_layer_model_complete_training_new on the GLUE QNLI dataset. It achieves the following results on the evaluation set:

• Loss: 0.6883
• Accuracy: 0.5416

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: 4e-05
• train_batch_size: 128
• eval_batch_size: 128
• 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

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6964	1.0	819	0.6927	0.5171
0.6946	2.0	1638	0.6927	0.4946
0.6936	3.0	2457	0.6915	0.5200
0.6998	4.0	3276	0.6902	0.4946
0.6925	5.0	4095	0.6933	0.5257
0.6917	6.0	4914	0.6893	0.5274
0.6914	7.0	5733	0.6894	0.5294
0.6916	8.0	6552	0.6888	0.5382
0.6913	9.0	7371	0.6883	0.5416
0.6909	10.0	8190	0.6892	0.5356
0.6914	11.0	9009	0.6892	0.5411
0.6918	12.0	9828	0.6907	0.5257
0.6911	13.0	10647	0.6905	0.5286
0.6909	14.0	11466	0.6896	0.5319

Framework versions

• Transformers 4.30.2
• Pytorch 1.14.0a0+410ce96
• Datasets 2.13.0
• Tokenizers 0.13.3