Downloads · 30 days
5
10% of all-time downloads
mlm
nomsgadded/mlm
mlm is a fill-mask model from nomsgadded. Use it when you need the model to fill a missing word. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
All-time downloads
49
Public
Parameters
125M
1.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors499 MB · 99%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- License
- mit
- Model type
- roberta
- Access
- Public
- Created
- Aug 28, 2023
- Updated
- Aug 28, 2023
- SHA
- b40530ef
Try a prompt
Base models
Datasets
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. -->
mlm
This model is a fine-tuned version of roberta-base on the wikitext wikitext-2-raw-v1 dataset. It achieves the following results on the evaluation set:
- Loss: 1.2468
- Accuracy: 0.7303
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: 8
- eval_batch_size: 8
- seed: 0
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.3758 | 1.0 | 150 | 1.2826 | 0.7277 |
| 1.3763 | 2.0 | 300 | 1.2747 | 0.7272 |
| 1.3558 | 3.0 | 450 | 1.2607 | 0.7278 |
Framework versions
- Transformers 4.33.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3
- Task
- Fill-Mask
- Library
- transformers
- Type
- roberta
- License
- mit
- Created
- Aug 28, 2023
- Updated
- Aug 28, 2023