Downloads · 30 days
7
5% of all-time downloads
bert-concat-2-finetune-simcse-truncate
NasimB/bert-concat-2-finetune-simcse-truncate
bert-concat-2-finetune-simcse-truncate is a fill-mask model from NasimB. 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
155
Public
Repo size
13 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin432 MB · 99%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- bert
- Access
- Public
- Created
- Jun 25, 2023
- Updated
- Jun 25, 2023
- SHA
- e366d4da
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. -->
bert-concat-2-finetune-simcse-truncate
This model was trained from scratch on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.7771
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: 0.0005
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.7122 | 0.25 | 1000 | 4.7219 |
| 4.8805 | 0.5 | 2000 | 4.1641 |
| 4.3911 | 0.76 | 3000 | 3.7771 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
- Task
- Fill-Mask
- Library
- transformers
- Type
- bert
- Created
- Jun 25, 2023
- Updated
- Jun 25, 2023