Downloads · 30 days
43
6% of all-time downloads
TinyBERT_L-4_H-312_v2-finetuned-wikitext103
saghar/TinyBERT_L-4_H-312_v2-finetuned-wikitext103
TinyBERT_L-4_H-312_v2-finetuned-wikitext103 is a fill-mask model from saghar. 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
723
Public
Repo size
1.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin57.6 MB · 99%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- bert
- Access
- Public
- Created
- Mar 17, 2022
- Updated
- Mar 17, 2022
- SHA
- a6f79a9b
Try a prompt
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. -->
TinyBERT_L-4_H-312_v2-finetuned-wikitext103
This model is a fine-tuned version of nreimers/TinyBERT_L-4_H-312_v2 on the wikitext dataset. It achieves the following results on the evaluation set:
- Loss: 6.4638
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.0604 | 1.0 | 3125 | 6.6745 |
| 6.7122 | 2.0 | 6250 | 6.5061 |
| 6.6289 | 3.0 | 9375 | 6.4638 |
Framework versions
- Transformers 4.16.2
- Pytorch 1.8.1
- Datasets 1.11.0
- Tokenizers 0.10.3
- Task
- Fill-Mask
- Library
- transformers
- Type
- bert
- Created
- Mar 17, 2022
- Updated
- Mar 17, 2022