Downloads · 30 days
7
21% of all-time downloads
variant-tapt_freeze_llrd-LR_2e-05
Mardiyyah/variant-tapt_freeze_llrd-LR_2e-05
variant-tapt_freeze_llrd-LR_2e-05 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
33
Public
Parameters
110M
884 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors442 MB · 100%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- License
- apache-2.0
- Model type
- bert
- Access
- Public
- Created
- Jan 5, 2026
- Updated
- Jan 6, 2026
- SHA
- db37f6b6
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. -->
variant-tapt_freeze_llrd-LR_2e-05
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on the Mardiyyah/TAPT_Variant_FT dataset. It achieves the following results on the evaluation set:
- Loss: 1.4408
- Accuracy: 0.7208
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-06 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.6327 | 1.0 | 19 | 1.6173 | 0.7107 |
| 1.5988 | 2.0 | 38 | 1.5820 | 0.7121 |
| 1.5615 | 3.0 | 57 | 1.5136 | 0.7221 |
| 1.5289 | 4.0 | 76 | 1.4800 | 0.7179 |
| 1.5178 | 5.0 | 95 | 1.4445 | 0.7289 |
| 1.4878 | 6.0 | 114 | 1.4360 | 0.7259 |
| 1.4812 | 7.0 | 133 | 1.4790 | 0.7208 |
| 1.4423 | 8.0 | 152 | 1.4385 | 0.7298 |
| 1.4869 | 9.0 | 171 | 1.4737 | 0.7273 |
| 1.4432 | 10.0 | 190 | 1.4679 | 0.7239 |
| 1.4259 | 11.0 | 209 | 1.3732 | 0.7280 |
| 1.4123 | 12.0 | 228 | 1.3996 | 0.7269 |
| 1.4365 | 13.0 | 247 | 1.3973 | 0.7317 |
| 1.4018 | 14.0 | 266 | 1.4079 | 0.7268 |
| 1.4105 | 15.0 | 285 | 1.3483 | 0.7309 |
| 1.409 | 16.0 | 304 | 1.4483 | 0.7189 |
| 1.4064 | 17.0 | 323 | 1.4226 | 0.7286 |
| 1.3731 | 18.0 | 342 | 1.3888 | 0.7275 |
| 1.4023 | 19.0 | 361 | 1.4027 | 0.7239 |
| 1.393 | 20.0 | 380 | 1.4187 | 0.7259 |
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
- Jan 5, 2026
- Updated
- Jan 6, 2026