bert-base-uncased_LOGIC_Native

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: 1.7437
• Accuracy: 0.61
• Macro Precision: 0.5758
• Macro F1: 0.5770

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro Precision	Macro F1
No log	1.0	116	2.3493	0.2433	0.3585	0.2208
No log	2.0	232	1.9408	0.4267	0.4271	0.3717
No log	3.0	348	1.6372	0.4733	0.4433	0.4383
No log	4.0	464	1.5114	0.59	0.5677	0.5507
1.7294	5.0	580	1.4359	0.6	0.5738	0.5761
1.7294	6.0	696	1.4476	0.6133	0.5913	0.5806
1.7294	7.0	812	1.4863	0.6133	0.5784	0.5823
1.7294	8.0	928	1.5494	0.63	0.5924	0.5903
0.3290	9.0	1044	1.6099	0.61	0.5733	0.5776
0.3290	10.0	1160	1.6845	0.6167	0.5809	0.5817
0.3290	11.0	1276	1.7364	0.6233	0.5811	0.5848
0.3290	12.0	1392	1.7437	0.61	0.5758	0.5770

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2