Downloads · 30 days
9
24% of all-time downloads
CeLLaTe-tapt-bioformer16l-tokenizer-adapted-spanmask-combinedData
Mardiyyah/CeLLaTe-tapt-bioformer16l-tokenizer-adapted-spanmask-combinedData
CeLLaTe-tapt-bioformer16l-tokenizer-adapted-spanmask-combinedData 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
37
Public
Parameters
42.5M
170 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors170 MB · 99%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- License
- apache-2.0
- Model type
- bert
- Access
- Public
- Created
- Jul 25, 2026
- Updated
- Jul 25, 2026
- SHA
- e055110c
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-bioformer16l-tokenizer-adapted-spanmask-combinedData
This model is a fine-tuned version of bioformers/bioformer-16l on the Mardiyyah/Combined_TAPT_dataset dataset. It achieves the following results on the evaluation set:
- Loss: 2.6024
- Accuracy: 0.5465
- Perplexity: 13.4958
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 |
|---|---|---|---|---|---|
| 3.552 | 1.0 | 30 | 3.0586 | 0.5154 | 21.2975 |
| 3.2568 | 2.0 | 60 | 2.9386 | 0.5208 | 18.8891 |
| 3.156 | 3.0 | 90 | 2.8878 | 0.5269 | 17.9534 |
| 3.0886 | 4.0 | 120 | 2.7759 | 0.5336 | 16.0535 |
| 3.0503 | 5.0 | 150 | 2.7636 | 0.5376 | 15.8571 |
| 3.0083 | 6.0 | 180 | 2.7304 | 0.5413 | 15.3387 |
| 2.9903 | 7.0 | 210 | 2.7123 | 0.5403 | 15.0639 |
| 2.9187 | 8.0 | 240 | 2.7158 | 0.5379 | 15.1165 |
| 2.8956 | 9.0 | 270 | 2.7472 | 0.5370 | 15.5986 |
| 2.8936 | 10.0 | 300 | 2.6687 | 0.5409 | 14.4215 |
| 2.8689 | 11.0 | 330 | 2.7531 | 0.5323 | 15.6912 |
| 2.8609 | 12.0 | 360 | 2.6994 | 0.5356 | 14.8709 |
| 2.8329 | 13.0 | 390 | 2.7289 | 0.5330 | 15.3156 |
| 2.7978 | 14.0 | 420 | 2.6334 | 0.5437 | 13.9212 |
| 2.7791 | 15.0 | 450 | 2.6627 | 0.5447 | 14.3353 |
| 2.761 | 16.0 | 480 | 2.6767 | 0.5382 | 14.5373 |
| 2.7705 | 17.0 | 510 | 2.6681 | 0.5394 | 14.4131 |
| 2.7463 | 18.0 | 540 | 2.6260 | 0.5416 | 13.8184 |
| 2.7437 | 19.0 | 570 | 2.6553 | 0.5397 | 14.2286 |
| 2.7234 | 20.0 | 600 | 2.6325 | 0.5441 | 13.9082 |
| 2.7091 | 21.0 | 630 | 2.5966 | 0.5487 | 13.4178 |
| 2.6875 | 22.0 | 660 | 2.6404 | 0.5395 | 14.0189 |
| 2.6903 | 23.0 | 690 | 2.6901 | 0.5337 | 14.7330 |
| 2.6605 | 24.0 | 720 | 2.5980 | 0.5484 | 13.4364 |
| 2.6802 | 25.0 | 750 | 2.6086 | 0.5378 | 13.5797 |
| 2.6641 | 26.0 | 780 | 2.6076 | 0.5424 | 13.5670 |
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 25, 2026
- Updated
- Jul 25, 2026