ULS-MultiClinNERsv-Qwen2.5-32B-symptom

This model is a fine-tuned version of Qwen/Qwen2.5-32B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0012
• Precision: 0.9983
• Recall: 0.9991
• F1: 0.9987
• Accuracy: 1.0000

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: 0.0002
• train_batch_size: 128
• eval_batch_size: 1
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	79	0.2487	0.2141	0.1506	0.1768	0.9058
No log	2.0	158	0.1746	0.3905	0.5030	0.4397	0.9392
No log	3.0	237	0.1016	0.5570	0.4978	0.5257	0.9631
No log	4.0	316	0.0589	0.6963	0.6823	0.6892	0.9796
No log	5.0	395	0.0336	0.8060	0.8642	0.8341	0.9904
No log	6.0	474	0.0197	0.9013	0.8903	0.8958	0.9939
0.2045	7.0	553	0.0081	0.9596	0.9713	0.9654	0.9981
0.2045	8.0	632	0.0028	0.9948	0.9974	0.9961	0.9998
0.2045	9.0	711	0.0015	0.9974	0.9983	0.9978	0.9999
0.2045	10.0	790	0.0012	0.9983	0.9991	0.9987	1.0000

Framework versions

• PEFT 0.14.0
• Transformers 4.47.0
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.0