bert_base_km_50_v2

This model is a fine-tuned version of on the Hartunka/processed_wikitext-103-raw-v1-km-50_v2 dataset. It achieves the following results on the evaluation set:

• Loss: 6.8821
• Accuracy: 0.1538

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: 0.0001
• train_batch_size: 96
• eval_batch_size: 96
• seed: 10
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10000
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
6.8758	4.1982	10000	6.9273	0.1497
6.4229	8.3963	20000	6.8691	0.1526
6.1843	12.5945	30000	6.9468	0.1552
6.0459	16.7926	40000	7.1213	0.1520
5.9642	20.9908	50000	7.0947	0.1527

Framework versions

• Transformers 4.50.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.21.1