bert-base-uncased-FinedTuned

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

• Loss: 2.7821
• Accuracy: 0.1762

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• training_steps: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2457	5.5556	1000	2.4913	0.1762
0.1759	11.1111	2000	2.8424	0.1762
0.1458	16.6667	3000	2.9765	0.1762
0.1132	22.2222	4000	2.7163	0.1762
0.1118	27.7778	5000	2.7272	0.1762
0.1272	33.3333	6000	2.8354	0.1762
0.1233	38.8889	7000	2.6948	0.1762
0.1161	44.4444	8000	2.7358	0.1762
0.0736	50.0	9000	2.7748	0.1762
0.0924	55.5556	10000	2.7821	0.1762

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1