BERT_BASE_TS_phonetic_wikitext_0.0

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: 1.7259

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: 1e-05
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10000
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.0301	0.3019	2000	1.8848
1.9624	0.6039	4000	1.8266
1.9287	0.9058	6000	1.7743
1.9046	1.2077	8000	1.7480
1.8906	1.5097	10000	1.7608
1.87	1.8116	12000	1.7273
1.8541	2.1135	14000	1.7327
1.8471	2.4155	16000	1.7083
1.8402	2.7174	18000	1.7417
1.8292	3.0193	20000	1.7046
1.8243	3.3213	22000	1.7018
1.8155	3.6232	24000	1.7189
1.8152	3.9251	26000	1.6891
1.8077	4.2271	28000	1.6809
1.811	4.5290	30000	1.6852
1.8118	4.8309	32000	1.7259

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0