LHL_LLM_Project

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

• Loss: 0.1958
• Accuracy: 0.925
• F1: 0.9247
• Precision: 0.9283
• Roc Auc: 0.6679

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: 7.578566728652807e-06
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 11

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Roc Auc
No log	1.0	391	0.2147	0.9160	0.9160	0.9161	0.6946
0.2956	2.0	782	0.1958	0.925	0.9247	0.9283	0.6679
0.1824	3.0	1173	0.2022	0.925	0.9262	0.9112	0.7734
0.1461	4.0	1564	0.2026	0.9277	0.9280	0.9237	0.7572
0.1461	5.0	1955	0.2331	0.9232	0.9251	0.9023	0.8164
0.1097	6.0	2346	0.2429	0.926	0.9271	0.9137	0.8139
0.0848	7.0	2737	0.2640	0.9234	0.9252	0.9040	0.8494
0.0707	8.0	3128	0.2622	0.9276	0.9281	0.9223	0.8237
0.0558	9.0	3519	0.2868	0.9256	0.9263	0.9173	0.8442
0.0558	10.0	3910	0.2926	0.9263	0.9271	0.9165	0.8514
0.0498	11.0	4301	0.2945	0.9264	0.9272	0.9176	0.8506

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu126
• Datasets 2.19.2
• Tokenizers 0.21.0