Downloads · 30 days
4
11% of all-time downloads
add_bert_12_layer_model_complete_training_new_96
gokuls/add_bert_12_layer_model_complete_training_new_96
add_bert_12_layer_model_complete_training_new_96 is a fill-mask model from gokuls. 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
35
Public
Repo size
9.7 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin467 MB · 89%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- hybridbert
- Access
- Public
- Created
- Jun 13, 2023
- Updated
- Jun 15, 2023
- SHA
- 23e52532
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. -->
add_bert_12_layer_model_complete_training_new_96
This model is a fine-tuned version of gokuls/add_bert_12_layer_model_complete_training_new_48 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 5.4112
- Accuracy: 0.1893
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: 1e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10000
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 5.8144 | 0.08 | 10000 | 5.7474 | 0.1593 |
| 5.7889 | 0.16 | 20000 | 5.7204 | 0.1604 |
| 5.6347 | 0.25 | 30000 | 5.6966 | 0.1623 |
| 5.7138 | 0.33 | 40000 | 5.6725 | 0.1636 |
| 5.6769 | 0.41 | 50000 | 5.6518 | 0.1658 |
| 5.6603 | 0.49 | 60000 | 5.6290 | 0.1686 |
| 5.5852 | 0.57 | 70000 | 5.6076 | 0.1707 |
| 5.6607 | 0.66 | 80000 | 5.5906 | 0.1720 |
| 5.5823 | 0.74 | 90000 | 5.5719 | 0.1739 |
| 5.6124 | 0.82 | 100000 | 5.5543 | 0.1759 |
| 5.6478 | 0.9 | 110000 | 5.5358 | 0.1776 |
| 5.4795 | 0.98 | 120000 | 5.5203 | 0.1787 |
| 5.4557 | 1.07 | 130000 | 5.5028 | 0.1804 |
| 5.5585 | 1.15 | 140000 | 5.4923 | 0.1814 |
| 5.6387 | 1.23 | 150000 | 5.4781 | 0.1825 |
| 5.479 | 1.31 | 160000 | 5.4663 | 0.1833 |
| 5.3951 | 1.39 | 170000 | 5.4512 | 0.1851 |
| 5.5062 | 1.47 | 180000 | 5.4411 | 0.1864 |
| 5.4553 | 1.56 | 190000 | 5.4244 | 0.1881 |
| 5.5461 | 1.64 | 200000 | 5.4112 | 0.1893 |
Framework versions
- Transformers 4.30.1
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
- Task
- Fill-Mask
- Library
- transformers
- Type
- hybridbert
- Created
- Jun 13, 2023
- Updated
- Jun 15, 2023