Downloads · 30 days
5
15% of all-time downloads
bert_12_layer_model_v2_complete_training
gokuls/bert_12_layer_model_v2_complete_training
bert_12_layer_model_v2_complete_training 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
34
Public
Repo size
9.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin210 MB · 99%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- hybridbert
- Access
- Public
- Created
- Feb 28, 2023
- Updated
- Mar 5, 2023
- SHA
- d0dabb94
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. -->
bert_12_layer_model_v2_complete_training
This model is a fine-tuned version of on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8623
- Accuracy: 0.6328
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: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- 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 |
|---|---|---|---|---|
| 6.1798 | 0.11 | 10000 | 6.1719 | 0.1485 |
| 6.0527 | 0.22 | 20000 | 6.0469 | 0.1502 |
| 5.6176 | 0.33 | 30000 | 5.5703 | 0.1772 |
| 3.8786 | 0.44 | 40000 | 3.7441 | 0.3851 |
| 3.4104 | 0.55 | 50000 | 3.3105 | 0.4327 |
| 3.1802 | 0.66 | 60000 | 3.0781 | 0.4601 |
| 3.0115 | 0.76 | 70000 | 2.9141 | 0.4804 |
| 2.8893 | 0.87 | 80000 | 2.7930 | 0.4956 |
| 2.7983 | 0.98 | 90000 | 2.6973 | 0.5081 |
| 2.7039 | 1.09 | 100000 | 2.6016 | 0.5215 |
| 2.5658 | 1.2 | 110000 | 2.4551 | 0.5448 |
| 2.4846 | 1.31 | 120000 | 2.3730 | 0.5576 |
| 2.4284 | 1.42 | 130000 | 2.3164 | 0.5663 |
| 2.3723 | 1.53 | 140000 | 2.2734 | 0.5726 |
| 2.3382 | 1.64 | 150000 | 2.2344 | 0.5787 |
| 2.3084 | 1.75 | 160000 | 2.2031 | 0.5829 |
| 2.2773 | 1.86 | 170000 | 2.1758 | 0.5872 |
| 2.2492 | 1.97 | 180000 | 2.1484 | 0.5909 |
| 2.2261 | 2.08 | 190000 | 2.1230 | 0.5943 |
| 2.1961 | 2.18 | 200000 | 2.1016 | 0.5976 |
| 2.1838 | 2.29 | 210000 | 2.0820 | 0.6004 |
| 2.164 | 2.4 | 220000 | 2.0645 | 0.6031 |
| 2.1456 | 2.51 | 230000 | 2.0469 | 0.6052 |
| 2.1308 | 2.62 | 240000 | 2.0293 | 0.6080 |
| 2.1161 | 2.73 | 250000 | 2.0137 | 0.6101 |
| 2.1052 | 2.84 | 260000 | 2.0020 | 0.6120 |
| 2.0856 | 2.95 | 270000 | 1.9902 | 0.6142 |
| 2.0743 | 3.06 | 280000 | 1.9775 | 0.6159 |
| 2.0598 | 3.17 | 290000 | 1.9678 | 0.6171 |
| 2.0492 | 3.28 | 300000 | 1.9561 | 0.6190 |
| 2.0395 | 3.39 | 310000 | 1.9453 | 0.6203 |
| 2.0328 | 3.5 | 320000 | 1.9365 | 0.6217 |
| 2.0204 | 3.6 | 330000 | 1.9287 | 0.6230 |
| 2.0142 | 3.71 | 340000 | 1.9199 | 0.6243 |
| 2.0021 | 3.82 | 350000 | 1.9121 | 0.6257 |
| 2.006 | 3.93 | 360000 | 1.9043 | 0.6264 |
| 1.9917 | 4.04 | 370000 | 1.8984 | 0.6274 |
| 1.9881 | 4.15 | 380000 | 1.8916 | 0.6284 |
| 1.9843 | 4.26 | 390000 | 1.8867 | 0.6291 |
| 1.977 | 4.37 | 400000 | 1.8809 | 0.6301 |
| 1.9697 | 4.48 | 410000 | 1.8770 | 0.6306 |
| 1.9655 | 4.59 | 420000 | 1.8740 | 0.6313 |
| 1.9649 | 4.7 | 430000 | 1.8691 | 0.6320 |
| 1.9622 | 4.81 | 440000 | 1.8662 | 0.6324 |
| 1.9539 | 4.92 | 450000 | 1.8623 | 0.6328 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.14.0a0+410ce96
- Datasets 2.10.1
- Tokenizers 0.13.2
- Task
- Fill-Mask
- Library
- transformers
- Type
- hybridbert
- Created
- Feb 28, 2023
- Updated
- Mar 5, 2023