mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_sst2

This model is a fine-tuned version of google/mobilebert-uncased on the GLUE SST2 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4834
• Accuracy: 0.8601

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	Accuracy
0.4923	1.0	8748	0.5804	0.8314
0.3226	2.0	17496	0.5184	0.8475
0.2725	3.0	26244	0.5341	0.8509
0.2453	4.0	34992	0.4892	0.8521
0.2278	5.0	43740	0.4834	0.8601
0.2149	6.0	52488	0.4980	0.8624
0.2047	7.0	61236	0.5031	0.8532
0.1963	8.0	69984	0.5011	0.8509
0.1893	9.0	78732	0.4899	0.8567
0.1835	10.0	87480	0.4965	0.8589

Framework versions

• Transformers 4.26.0
• Pytorch 1.14.0a0+410ce96
• Datasets 2.9.0
• Tokenizers 0.13.2