Downloads · 30 days
12
2% of all-time downloads
roberta-base-finetuned-wikitext2
graphcore-rahult/roberta-base-finetuned-wikitext2
roberta-base-finetuned-wikitext2 is a fill-mask model from graphcore-rahult. 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
504
Public
Repo size
748 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin249 MB · 100%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- License
- mit
- Model type
- roberta
- Access
- Public
- Created
- Dec 3, 2022
- Updated
- Dec 6, 2022
- SHA
- 9fc667cd
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. -->
roberta-base-finetuned-wikitext2
This model is a fine-tuned version of roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.5068
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: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: IPU
- gradient_accumulation_steps: 16
- total_train_batch_size: 64
- total_eval_batch_size: 20
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
- training precision: Mixed Precision
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.571 | 1.0 | 300 | 1.5332 |
| 1.5725 | 2.0 | 600 | 1.5117 |
| 1.4742 | 3.0 | 900 | 1.5078 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.10.0+cpu
- Datasets 2.7.1
- Tokenizers 0.12.1
- Task
- Fill-Mask
- Library
- transformers
- Type
- roberta
- License
- mit
- Created
- Dec 3, 2022
- Updated
- Dec 6, 2022