scenario-KD-SCR-DIV2-data-glue-sst2-model-bert-base-uncased-run-3

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.8394
• Accuracy: 0.8911

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: 32
• eval_batch_size: 32
• seed: 45
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 6969

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.379	1.0	2105	1.9908	0.8784
0.8041	2.0	4210	1.6913	0.8865
0.6181	3.0	6315	1.9046	0.8784
0.4617	4.0	8420	1.6333	0.8922
0.41	5.0	10525	1.8034	0.8830
0.351	6.0	12630	1.8450	0.8899
0.32	7.0	14735	1.9411	0.8842
0.2927	8.0	16840	1.7726	0.8991
0.2771	9.0	18945	1.6832	0.8807
0.2565	10.0	21050	1.6899	0.8933
0.2434	11.0	23155	1.7471	0.8888
0.23	12.0	25260	2.1582	0.8704
0.2255	13.0	27365	1.8486	0.9037
0.2041	14.0	29470	1.6787	0.8922
0.218	15.0	31575	1.8426	0.8888
0.2027	16.0	33680	1.8820	0.8945
0.1924	17.0	35785	1.9447	0.8933
0.1964	18.0	37890	1.8394	0.8911

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2
• Datasets 2.16.0
• Tokenizers 0.15.0