|
|
--- |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- generator |
|
|
model-index: |
|
|
- name: bert-concat-2 |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- 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 |
|
|
|
|
|
This model is a fine-tuned version of [](https://huggingface.co/) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 5.7060 |
|
|
|
|
|
## 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 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: cosine |
|
|
- lr_scheduler_warmup_steps: 1000 |
|
|
- num_epochs: 20 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
|
| 6.6866 | 0.52 | 1000 | 6.2709 | |
|
|
| 6.2315 | 1.04 | 2000 | 6.2177 | |
|
|
| 6.1818 | 1.56 | 3000 | 6.1895 | |
|
|
| 6.1511 | 2.08 | 4000 | 6.1559 | |
|
|
| 6.0984 | 2.6 | 5000 | 6.1185 | |
|
|
| 6.0611 | 3.12 | 6000 | 6.0668 | |
|
|
| 6.0114 | 3.65 | 7000 | 6.0361 | |
|
|
| 5.9679 | 4.17 | 8000 | 6.0160 | |
|
|
| 5.9272 | 4.69 | 9000 | 5.9731 | |
|
|
| 5.8904 | 5.21 | 10000 | 5.9424 | |
|
|
| 5.8557 | 5.73 | 11000 | 5.9190 | |
|
|
| 5.8237 | 6.25 | 12000 | 5.9002 | |
|
|
| 5.8008 | 6.77 | 13000 | 5.8787 | |
|
|
| 5.7785 | 7.29 | 14000 | 5.8644 | |
|
|
| 5.7569 | 7.81 | 15000 | 5.8534 | |
|
|
| 5.7305 | 8.33 | 16000 | 5.8429 | |
|
|
| 5.7187 | 8.85 | 17000 | 5.8283 | |
|
|
| 5.699 | 9.38 | 18000 | 5.8124 | |
|
|
| 5.6737 | 9.9 | 19000 | 5.8055 | |
|
|
| 5.648 | 10.42 | 20000 | 5.7945 | |
|
|
| 5.641 | 10.94 | 21000 | 5.7869 | |
|
|
| 5.613 | 11.46 | 22000 | 5.7700 | |
|
|
| 5.6078 | 11.98 | 23000 | 5.7659 | |
|
|
| 5.5759 | 12.5 | 24000 | 5.7555 | |
|
|
| 5.5682 | 13.02 | 25000 | 5.7522 | |
|
|
| 5.5461 | 13.54 | 26000 | 5.7397 | |
|
|
| 5.5414 | 14.06 | 27000 | 5.7349 | |
|
|
| 5.5195 | 14.58 | 28000 | 5.7310 | |
|
|
| 5.5081 | 15.1 | 29000 | 5.7214 | |
|
|
| 5.4922 | 15.62 | 30000 | 5.7188 | |
|
|
| 5.4858 | 16.15 | 31000 | 5.7127 | |
|
|
| 5.4786 | 16.67 | 32000 | 5.7092 | |
|
|
| 5.4685 | 17.19 | 33000 | 5.7075 | |
|
|
| 5.4571 | 17.71 | 34000 | 5.7060 | |
|
|
| 5.4592 | 18.23 | 35000 | 5.7018 | |
|
|
| 5.4555 | 18.75 | 36000 | 5.7043 | |
|
|
| 5.4512 | 19.27 | 37000 | 5.7028 | |
|
|
| 5.4522 | 19.79 | 38000 | 5.7060 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.26.1 |
|
|
- Pytorch 1.11.0+cu113 |
|
|
- Datasets 2.13.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
