ClinicalBERT-mimic-phi-ner

racheltong/ClinicalBERT-mimic-phi-ner

ClinicalBERT-mimic-phi-ner is a token classification model from racheltong. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as mit.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

2% of all-time downloads

All-time downloads

425

Public

Parameters

108M

27.6 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt862 MB · 50%

At a glance

Task: Token Classification
Library: transformers
License: mit
Model type: bert
Access: Public
Created: Mar 26, 2026
Updated: Mar 26, 2026
SHA: 738c325b

Try a prompt

Base models

emilyalsentzer/Bio_ClinicalBERT

Model card

From the Hugging Face model README

ClinicalBERT-mimic-phi-ner

This model is a fine-tuned version of emilyalsentzer/Bio_ClinicalBERT on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0017
F1 Macro: 0.9441
F1 Weighted: 0.9441
Precision: 0.9140
Recall: 0.9763
F1 Name: 0.94
F1 Location: 0.91
F1 Phone: 0.93
F1 Date: 0.84
F1 Mrn: 0.96
F1 Account: 0.97
F1 Age Over 89: 0.98
F1 Device Id: 0.99
F1 Ssn: 1.0
F1 Url: 1.0
F1 Email: 0.99

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: 32
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
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: cosine
lr_scheduler_warmup_steps: 0.1
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Macro	F1 Weighted	Precision	Recall	F1 Name	F1 Location	F1 Phone	F1 Date	F1 Mrn	F1 Account	F1 Age Over 89	F1 Device Id	F1 Ssn	F1 Url	F1 Email
0.4470	0.1774	300	0.0868	0.3948	0.3948	0.2935	0.6032	0.46	0.33	0.41	0.04	0.46	0.4	0.08	0.58	0.32	0.0	0.31
0.0508	0.3547	600	0.0112	0.7449	0.7449	0.6654	0.8461	0.82	0.57	0.8	0.21	0.64	0.85	0.04	0.89	0.86	0.56	0.95
0.0302	0.5321	900	0.0131	0.8389	0.8389	0.7652	0.9284	0.88	0.72	0.86	0.27	0.59	0.98	0.84	0.9	0.92	0.93	0.99
0.0244	0.7094	1200	0.0046	0.8816	0.8816	0.8212	0.9517	0.9	0.81	0.81	0.48	0.75	0.97	0.95	0.97	0.98	1.0	1.0
0.0187	0.8868	1500	0.0030	0.9160	0.9160	0.8713	0.9656	0.93	0.82	0.87	0.52	0.89	0.95	0.96	0.96	1.0	1.0	1.0
0.0055	1.0638	1800	0.0030	0.9343	0.9343	0.8979	0.9737	0.94	0.89	0.9	0.57	0.92	0.97	0.98	0.99	1.0	1.0	1.0
0.0037	1.2412	2100	0.0027	0.9306	0.9306	0.8944	0.9697	0.93	0.89	0.9	0.74	0.92	0.98	0.98	0.99	1.0	1.0	1.0
0.0117	1.4186	2400	0.0025	0.9338	0.9338	0.8988	0.9716	0.94	0.88	0.89	0.8	0.94	0.97	0.98	0.99	1.0	1.0	0.99
0.0066	1.5959	2700	0.0020	0.9454	0.9454	0.9159	0.9769	0.95	0.9	0.93	0.83	0.96	0.98	0.99	0.99	1.0	1.0	0.99
0.0043	1.7733	3000	0.0018	0.9433	0.9433	0.9124	0.9763	0.94	0.9	0.93	0.82	0.96	0.97	0.99	0.99	1.0	1.0	0.99
0.0030	1.9506	3300	0.0017	0.9441	0.9441	0.9140	0.9763	0.94	0.91	0.93	0.84	0.96	0.97	0.98	0.99	1.0	1.0	0.99

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

Task: Token Classification
Library: transformers
Type: bert
License: mit
Created: Mar 26, 2026
Updated: Mar 26, 2026