Downloads · 30 days
10
3% of all-time downloads
tiny-mlm-snli-from-scratch
muhtasham/tiny-mlm-snli-from-scratch
tiny-mlm-snli-from-scratch is a fill-mask model from muhtasham. Use it when you need the model to fill a missing word. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
335
Public
Repo size
195 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin17.7 MB · 95%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- bert
- Access
- Public
- Created
- Jan 20, 2023
- Updated
- Jan 20, 2023
- SHA
- 731e5fb5
Try a prompt
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. -->
tiny-mlm-snli-from-scratch
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
- Loss: 5.6898
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: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 9.0745 | 0.4 | 500 | 7.9050 |
| 7.0157 | 0.8 | 1000 | 6.5279 |
| 6.0761 | 1.2 | 1500 | 6.0836 |
| 5.8313 | 1.6 | 2000 | 6.0346 |
| 5.7371 | 2.0 | 2500 | 5.9809 |
| 5.6413 | 2.4 | 3000 | 5.8541 |
| 5.5216 | 2.8 | 3500 | 5.8338 |
| 5.4991 | 3.2 | 4000 | 5.7308 |
| 5.4286 | 3.6 | 4500 | 5.6831 |
| 5.3192 | 4.0 | 5000 | 5.6898 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2
- Task
- Fill-Mask
- Library
- transformers
- Type
- bert
- Created
- Jan 20, 2023
- Updated
- Jan 20, 2023