BERTModified-rawbert-finetuned-wikitext-test

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

• Loss: 20.8186
• Precision: 0.0476
• Recall: 0.0476
• F1: 0.0476
• Accuracy: 0.0476

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
21.0846	1.0	25	20.9953	0.0114	0.0114	0.0114	0.0114
17.8286	2.0	50	20.7823	0.0114	0.0114	0.0114	0.0114
15.1916	3.0	75	20.7021	0.0171	0.0171	0.0171	0.0171
12.7015	4.0	100	20.6023	0.0248	0.0248	0.0248	0.0248
10.852	5.0	125	20.5528	0.0324	0.0324	0.0324	0.0324
9.2624	6.0	150	20.5556	0.0324	0.0324	0.0324	0.0324
7.8348	7.0	175	20.5343	0.0343	0.0343	0.0343	0.0343
6.762	8.0	200	20.5861	0.0381	0.0381	0.0381	0.0381
5.8667	9.0	225	20.6005	0.0381	0.0381	0.0381	0.0381
5.184	10.0	250	20.6594	0.0438	0.0438	0.0438	0.0438
4.4605	11.0	275	20.6880	0.0457	0.0457	0.0457	0.0457
4.106	12.0	300	20.7090	0.0457	0.0457	0.0457	0.0457
3.622	13.0	325	20.7341	0.0457	0.0457	0.0457	0.0457
3.3097	14.0	350	20.7556	0.0476	0.0476	0.0476	0.0476
3.0423	15.0	375	20.8040	0.0495	0.0495	0.0495	0.0495
2.8348	16.0	400	20.8144	0.0533	0.0533	0.0533	0.0533
2.6718	17.0	425	20.8144	0.0495	0.0495	0.0495	0.0495
2.5584	18.0	450	20.8312	0.0533	0.0533	0.0533	0.0533
2.4502	19.0	475	20.8228	0.0514	0.0514	0.0514	0.0514
2.4219	20.0	500	20.8194	0.0514	0.0514	0.0514	0.0514

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2