Downloads · 30 days
25
32% of all-time downloads
CeLLaTe-tapt-bioformer8l-tokenizer-original-wwmask
Mardiyyah/CeLLaTe-tapt-bioformer8l-tokenizer-original-wwmask
CeLLaTe-tapt-bioformer8l-tokenizer-original-wwmask 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
79
Public
Parameters
42.6M
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
- 4633c728
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-bioformer8l-tokenizer-original-wwmask
This model is a fine-tuned version of bioformers/bioformer-8l on the Mardiyyah/TAPT_CeLLaTe2.0 dataset. It achieves the following results on the evaluation set:
- Loss: 1.3721
- Accuracy: 0.7194
- Perplexity: 3.9435
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 |
|---|---|---|---|---|---|
| 1.6365 | 1.0 | 15 | 1.4316 | 0.7114 | 4.1855 |
| 1.6272 | 2.0 | 30 | 1.4087 | 0.7145 | 4.0906 |
| 1.6169 | 3.0 | 45 | 1.4094 | 0.7117 | 4.0936 |
| 1.5972 | 4.0 | 60 | 1.3890 | 0.7185 | 4.0110 |
| 1.5757 | 5.0 | 75 | 1.3756 | 0.7159 | 3.9575 |
| 1.5417 | 6.0 | 90 | 1.3728 | 0.7231 | 3.9463 |
| 1.5503 | 7.0 | 105 | 1.3709 | 0.7189 | 3.9387 |
| 1.5114 | 8.0 | 120 | 1.3977 | 0.7135 | 4.0458 |
| 1.4794 | 9.0 | 135 | 1.3369 | 0.7228 | 3.8074 |
| 1.4929 | 10.0 | 150 | 1.3787 | 0.7150 | 3.9698 |
| 1.4716 | 11.0 | 165 | 1.3592 | 0.7208 | 3.8930 |
| 1.4746 | 12.0 | 180 | 1.3830 | 0.7141 | 3.9869 |
| 1.4506 | 13.0 | 195 | 1.3438 | 0.7212 | 3.8337 |
| 1.4481 | 14.0 | 210 | 1.3879 | 0.7131 | 4.0064 |
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