|
|
--- |
|
|
license: mit |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- generator |
|
|
model-index: |
|
|
- name: all-base5 |
|
|
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. --> |
|
|
|
|
|
# all-base5 |
|
|
|
|
|
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 4.0357 |
|
|
|
|
|
## 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: 6 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
|
| 6.3501 | 0.29 | 500 | 5.3140 | |
|
|
| 5.0366 | 0.58 | 1000 | 4.8996 | |
|
|
| 4.7014 | 0.87 | 1500 | 4.6570 | |
|
|
| 4.4484 | 1.16 | 2000 | 4.5100 | |
|
|
| 4.2883 | 1.45 | 2500 | 4.3964 | |
|
|
| 4.1912 | 1.74 | 3000 | 4.2890 | |
|
|
| 4.0845 | 2.02 | 3500 | 4.2065 | |
|
|
| 3.8854 | 2.31 | 4000 | 4.1673 | |
|
|
| 3.8583 | 2.6 | 4500 | 4.1114 | |
|
|
| 3.816 | 2.89 | 5000 | 4.0592 | |
|
|
| 3.6485 | 3.18 | 5500 | 4.0531 | |
|
|
| 3.5744 | 3.47 | 6000 | 4.0243 | |
|
|
| 3.5619 | 3.76 | 6500 | 3.9912 | |
|
|
| 3.4959 | 4.05 | 7000 | 3.9799 | |
|
|
| 3.3058 | 4.34 | 7500 | 3.9807 | |
|
|
| 3.2993 | 4.63 | 8000 | 3.9659 | |
|
|
| 3.2879 | 4.92 | 8500 | 3.9544 | |
|
|
| 3.1683 | 5.21 | 9000 | 3.9630 | |
|
|
| 3.1172 | 5.49 | 9500 | 3.9624 | |
|
|
| 3.118 | 5.78 | 10000 | 3.9614 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.26.1 |
|
|
- Pytorch 1.11.0+cu113 |
|
|
- Datasets 2.13.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
