bert-large-uncased-sst-2-32-13-30

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

• Loss: 0.6452
• Accuracy: 0.6719

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: 1.5e-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: linear
• lr_scheduler_warmup_steps: 5
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	2	0.7180	0.4688
No log	2.0	4	0.7071	0.4688
No log	3.0	6	0.6996	0.5469
No log	4.0	8	0.6827	0.5625
0.6456	5.0	10	0.6712	0.5469
0.6456	6.0	12	0.6542	0.6094
0.6456	7.0	14	0.6525	0.6719
0.6456	8.0	16	0.6535	0.6875
0.6456	9.0	18	0.6454	0.6406
0.4153	10.0	20	0.6414	0.625
0.4153	11.0	22	0.6470	0.6094
0.4153	12.0	24	0.6509	0.6094
0.4153	13.0	26	0.6489	0.6094
0.4153	14.0	28	0.6498	0.6094
0.238	15.0	30	0.6514	0.6094
0.238	16.0	32	0.6440	0.6562
0.238	17.0	34	0.6432	0.6719
0.238	18.0	36	0.6497	0.6719
0.238	19.0	38	0.6569	0.6406
0.1523	20.0	40	0.6636	0.6094
0.1523	21.0	42	0.6692	0.5781
0.1523	22.0	44	0.6740	0.5625
0.1523	23.0	46	0.6708	0.5625
0.1523	24.0	48	0.6632	0.6094
0.1187	25.0	50	0.6596	0.6406
0.1187	26.0	52	0.6560	0.6562
0.1187	27.0	54	0.6517	0.6719
0.1187	28.0	56	0.6482	0.6875
0.1187	29.0	58	0.6462	0.6875
0.09	30.0	60	0.6452	0.6719

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.4.0
• Tokenizers 0.13.3