mobilebert_add_GLUE_Experiment_cola_256

This model is a fine-tuned version of google/mobilebert-uncased on the GLUE COLA dataset. It achieves the following results on the evaluation set:

• Loss: 0.6102
• Matthews Correlation: 0.0185

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: 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	Matthews Correlation
0.6129	1.0	67	0.6180	0.0
0.6078	2.0	134	0.6178	0.0
0.6073	3.0	201	0.6179	0.0
0.6067	4.0	268	0.6167	0.0
0.6059	5.0	335	0.6168	0.0
0.5998	6.0	402	0.6115	0.0
0.5917	7.0	469	0.6122	0.0
0.5849	8.0	536	0.6126	0.0
0.5796	9.0	603	0.6277	0.0
0.5759	10.0	670	0.6138	0.0029
0.5733	11.0	737	0.6102	0.0185
0.5716	12.0	804	0.6143	0.0252
0.5667	13.0	871	0.6347	0.0348
0.5662	14.0	938	0.6314	0.0385
0.5631	15.0	1005	0.6130	0.0174
0.5628	16.0	1072	0.6218	0.0348

Framework versions

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