Downloads · 30 days
22
8% of all-time downloads
arxiv-new-datasets-modernbert-v4
librarian-bots/arxiv-new-datasets-modernbert-v4
arxiv-new-datasets-modernbert-v4 is a text classification model from librarian-bots. Use it when you need a label for a piece of text. 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
289
Public
Parameters
150M
2.4 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors598 MB · 99%
At a glance
- Task
- Text Classification
- Library
- transformers
- License
- apache-2.0
- Model type
- modernbert
- Access
- Public
- Created
- Feb 7, 2026
- Updated
- Feb 7, 2026
- SHA
- fae36e1c
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. -->
arxiv-new-datasets-modernbert-v4
This model is a fine-tuned version of answerdotai/ModernBERT-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2915
- Accuracy: 0.9563
- F1: 0.9564
- Precision: 0.9569
- Recall: 0.9563
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: 42
- 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: 0.1
- num_epochs: 10
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.2884 | 1.0 | 425 | 0.3036 | 0.9430 | 0.9430 | 0.9430 | 0.9430 |
| 0.2557 | 2.0 | 850 | 0.2915 | 0.9563 | 0.9564 | 0.9569 | 0.9563 |
| 0.2247 | 3.0 | 1275 | 0.3049 | 0.9510 | 0.9506 | 0.9513 | 0.9510 |
| 0.2007 | 4.0 | 1700 | 0.3002 | 0.9550 | 0.9550 | 0.9551 | 0.9550 |
Framework versions
- Transformers 4.48.0.dev0
- Pytorch 2.10.0+cu128
- Datasets 4.5.0
- Tokenizers 0.21.4
- Task
- Text Classification
- Library
- transformers
- Type
- modernbert
- License
- apache-2.0
- Created
- Feb 7, 2026
- Updated
- Feb 7, 2026