CeLLaTe-ner-3class-bioformer16l-tapt-tokenizer-adapted-wwmask

Mardiyyah/CeLLaTe-ner-3class-bioformer16l-tapt-tokenizer-adapted-wwmask

CeLLaTe-ner-3class-bioformer16l-tapt-tokenizer-adapted-wwmask is a token classification model from Mardiyyah. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as apache-2.0.

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

Downloads · 30 days

15% of all-time downloads

All-time downloads

Public

Parameters

42.3M

169 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors169 MB · 99%

At a glance

Task: Token Classification
Library: transformers
License: apache-2.0
Model type: bert
Access: Public
Created: Jul 25, 2026
Updated: Jul 25, 2026
SHA: 7ddc17a2

Try a prompt

Base models

Mardiyyah/CeLLaTe-tapt-bioformer16l-tokenizer-adapted-wwmask

Model card

From the Hugging Face model README

CeLLaTe-ner-3class-bioformer16l-tapt-tokenizer-adapted-wwmask

This model is a fine-tuned version of Mardiyyah/CeLLaTe-tapt-bioformer16l-tokenizer-adapted-wwmask on the OTAR3088/CeLLaTe-ner-3class-iob_final dataset. It achieves the following results on the evaluation set:

Loss: 0.1079
Precision: 0.8003
Recall: 0.7357
Micro F1: 0.7667
Weighted F1: 0.7649
Macro F1: 0.7527
Accuracy: 0.9810

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: 3407
gradient_accumulation_steps: 2
total_train_batch_size: 32
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
lr_scheduler_warmup_ratio: 0.01
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	Micro F1	Weighted F1	Macro F1	Accuracy
0.402	1.0	263	0.1010	0.5231	0.5925	0.5557	0.5484	0.5226	0.9693
0.0736	2.0	526	0.0762	0.7704	0.6977	0.7322	0.7298	0.7167	0.9795
0.0455	3.0	789	0.0828	0.6888	0.6442	0.6657	0.6668	0.6607	0.9753
0.0345	4.0	1052	0.0824	0.7998	0.7163	0.7558	0.7528	0.7390	0.9803
0.0263	5.0	1315	0.0813	0.7303	0.7218	0.7260	0.7264	0.7173	0.9783
0.021	6.0	1578	0.0885	0.6907	0.6985	0.6946	0.6968	0.6894	0.9766
0.0171	7.0	1841	0.0936	0.7847	0.7144	0.7479	0.7465	0.7341	0.9799
0.0143	8.0	2104	0.0993	0.7401	0.6962	0.7175	0.7180	0.7087	0.9784
0.0115	9.0	2367	0.1013	0.7932	0.7280	0.7592	0.7574	0.7442	0.9806
0.0103	10.0	2630	0.1043	0.7520	0.7156	0.7333	0.7330	0.7224	0.9791
0.0095	11.0	2893	0.1053	0.7852	0.7346	0.7590	0.7572	0.7455	0.9807
0.0079	12.0	3156	0.1083	0.8003	0.7357	0.7667	0.7649	0.7527	0.9810
0.0067	13.0	3419	0.1095	0.7628	0.7264	0.7442	0.7432	0.7323	0.9796
0.0063	14.0	3682	0.1104	0.7727	0.7268	0.7491	0.7485	0.7372	0.9800
0.0059	15.0	3945	0.1123	0.7790	0.7416	0.7598	0.7597	0.7488	0.9801
0.0052	16.0	4208	0.1154	0.7825	0.7385	0.7598	0.7580	0.7456	0.9804
0.0048	17.0	4471	0.1179	0.7796	0.7342	0.7562	0.7555	0.7451	0.9801

Framework versions

Transformers 4.48.2
Pytorch 2.4.1+cu121
Datasets 3.0.2
Tokenizers 0.21.0

Task: Token Classification
Library: transformers
Type: bert
License: apache-2.0
Languages: en
Created: Jul 25, 2026
Updated: Jul 25, 2026