bert-large-uncased-sst-2-64-13-30

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: 0.6366
• Accuracy: 0.8594

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	4	0.6822	0.5781
No log	2.0	8	0.6592	0.6172
0.6625	3.0	12	0.6330	0.6719
0.6625	4.0	16	0.5916	0.75
0.5388	5.0	20	0.5256	0.8047
0.5388	6.0	24	0.4758	0.8125
0.5388	7.0	28	0.4535	0.8203
0.3481	8.0	32	0.4449	0.8203
0.3481	9.0	36	0.4246	0.8203
0.2064	10.0	40	0.3979	0.8359
0.2064	11.0	44	0.3869	0.8594
0.2064	12.0	48	0.3902	0.8672
0.1105	13.0	52	0.4068	0.8516
0.1105	14.0	56	0.4299	0.8516
0.0583	15.0	60	0.4556	0.8516
0.0583	16.0	64	0.4352	0.8594
0.0583	17.0	68	0.4515	0.8672
0.0313	18.0	72	0.4882	0.8672
0.0313	19.0	76	0.5242	0.8516
0.0201	20.0	80	0.5497	0.8594
0.0201	21.0	84	0.5667	0.8594
0.0201	22.0	88	0.5827	0.8594
0.0132	23.0	92	0.6062	0.8516
0.0132	24.0	96	0.6265	0.8516
0.017	25.0	100	0.6330	0.8438
0.017	26.0	104	0.6334	0.8516
0.017	27.0	108	0.6354	0.8516
0.0211	28.0	112	0.6361	0.8594
0.0211	29.0	116	0.6363	0.8594
0.0113	30.0	120	0.6366	0.8594

Framework versions

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