distilbert-base-uncased-finetuned-final

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4175
• Precision: 0.8939

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: 64
• eval_batch_size: 64
• 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
• training_steps: 240

Training results

Training Loss	Epoch	Step	Validation Loss	Precision
0.041	0.1271	15	0.5138	0.9186
0.0264	0.2542	30	0.4652	0.9064
0.0666	0.3814	45	0.4839	0.9137
0.0318	0.5085	60	0.4048	0.8930
0.0174	0.6356	75	0.4200	0.8797
0.0482	0.7627	90	0.5550	0.8514
0.0338	0.8898	105	0.5001	0.9160
0.0166	1.0169	120	0.4397	0.8908
0.022	1.1441	135	0.4291	0.8979
0.0055	1.2712	150	0.4433	0.8990
0.026	1.3983	165	0.5158	0.8689
0.0278	1.5254	180	0.4490	0.9019
0.0736	1.6525	195	0.4308	0.8925
0.0306	1.7797	210	0.4223	0.8925
0.0473	1.9068	225	0.4191	0.8932
0.0428	2.0339	240	0.4175	0.8939

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.1+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1