|
|
--- |
|
|
base_model: avichr/heBERT |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: heBERT-finetuned-samaritan |
|
|
results: [] |
|
|
widget: |
|
|
- text: "אלין שמהת בני ישראל דעלו למצרים עם [MASK] גבר וביתה עלו" |
|
|
- text: "בראשית ברא [MASK] את השמים ואת הארץ" |
|
|
- text: "ומשה הוה רעה ית עאן [MASK] חתנה כהן מדין ודעק ית עאנה אחרי מדברה ואתא לטור האלהים לחורב" |
|
|
- text: "וחזה לה [MASK] יהוה בלהבת אש מבגו סניה וחזה ואה סניה יקיד באש וסניה ליתו מתאכל" |
|
|
--- |
|
|
|
|
|
<!-- 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. --> |
|
|
|
|
|
# heBERT-finetuned-samaritan |
|
|
|
|
|
This model is a fine-tuned version of [avichr/heBERT](https://huggingface.co/avichr/heBERT) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 4.0412 |
|
|
|
|
|
## 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: 64 |
|
|
- eval_batch_size: 64 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 10 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:----:|:---------------:| |
|
|
| 4.875 | 1.0 | 118 | 4.5078 | |
|
|
| 4.5415 | 2.0 | 236 | 4.4287 | |
|
|
| 4.3991 | 3.0 | 354 | 4.2731 | |
|
|
| 4.3072 | 4.0 | 472 | 4.2191 | |
|
|
| 4.2421 | 5.0 | 590 | 4.1209 | |
|
|
| 4.2013 | 6.0 | 708 | 4.0951 | |
|
|
| 4.1624 | 7.0 | 826 | 4.0689 | |
|
|
| 4.1062 | 8.0 | 944 | 4.0910 | |
|
|
| 4.1 | 9.0 | 1062 | 4.0374 | |
|
|
| 4.0738 | 10.0 | 1180 | 4.0750 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.40.1 |
|
|
- Pytorch 2.3.0+cu121 |
|
|
- Datasets 2.19.0 |
|
|
- Tokenizers 0.19.1 |
|
|
|
