Downloads · 30 days
33
29% of all-time downloads
CeLLaTe-tapt-pubmedbert-tokenizer-original-baseline
Mardiyyah/CeLLaTe-tapt-pubmedbert-tokenizer-original-baseline
CeLLaTe-tapt-pubmedbert-tokenizer-original-baseline is a fill-mask model from Mardiyyah. Use it when you need the model to fill a missing word. 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. --
All-time downloads
114
Public
Parameters
110M
19.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt12.4 GB · 64%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- License
- apache-2.0
- Model type
- bert
- Access
- Public
- Created
- Jul 23, 2026
- Updated
- Jul 25, 2026
- SHA
- 9bd97fc0
Try a prompt
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
CeLLaTe-tapt-pubmedbert-tokenizer-original-baseline
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on the Mardiyyah/TAPT_CeLLaTe2.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.9726
- Accuracy: 0.7783
- Perplexity: 2.6448
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 3407
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 100
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Perplexity |
|---|---|---|---|---|---|
| 1.1444 | 1.0 | 14 | 1.0112 | 0.7749 | 2.7489 |
| 1.1264 | 2.0 | 28 | 0.9842 | 0.7806 | 2.6757 |
| 1.118 | 3.0 | 42 | 0.9774 | 0.7810 | 2.6574 |
| 1.0848 | 4.0 | 56 | 0.9892 | 0.7781 | 2.6891 |
| 1.0813 | 5.0 | 70 | 0.9697 | 0.7861 | 2.6372 |
| 1.0682 | 6.0 | 84 | 0.9910 | 0.7782 | 2.6938 |
| 1.0576 | 7.0 | 98 | 0.9641 | 0.7807 | 2.6225 |
| 1.0377 | 8.0 | 112 | 0.9392 | 0.7880 | 2.5580 |
| 1.0206 | 9.0 | 126 | 0.9936 | 0.7765 | 2.7009 |
| 1.0254 | 10.0 | 140 | 0.9782 | 0.7797 | 2.6595 |
| 0.9976 | 11.0 | 154 | 0.9627 | 0.7833 | 2.6188 |
| 0.9818 | 12.0 | 168 | 0.9549 | 0.7833 | 2.5985 |
| 0.9706 | 13.0 | 182 | 0.9498 | 0.7855 | 2.5851 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.21.0
- Task
- Fill-Mask
- Library
- transformers
- Type
- bert
- License
- apache-2.0
- Languages
- en
- Created
- Jul 23, 2026
- Updated
- Jul 25, 2026