bert-base-uncased-multi-128

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.7101

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss
3.6636	1.0	812	3.2325
3.2963	2.0	1624	3.1937
3.1132	3.0	2436	3.2984
2.9386	4.0	3248	3.2430
2.7742	5.0	4060	3.1272
2.5954	6.0	4872	3.1778
2.501	7.0	5684	3.1649
2.4073	8.0	6496	2.9395
2.2933	9.0	7308	3.1262
2.2218	10.0	8120	2.9994
2.1558	11.0	8932	2.9922
2.0873	12.0	9744	2.8414
2.0104	13.0	10556	2.9351
1.9364	14.0	11368	2.9253
1.9045	15.0	12180	2.8701
1.9152	16.0	12992	2.7101

Framework versions

• Transformers 4.8.2
• Pytorch 1.7.1
• Datasets 1.16.1
• Tokenizers 0.10.3