|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: gpt2-dp-guten-rarity-all-5k-2p5k |
|
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. --> |
|
|
|
# gpt2-dp-guten-rarity-all-5k-2p5k |
|
|
|
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: 4.3172 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 6.6951 | 0.28 | 500 | 5.6610 | |
|
| 5.3498 | 0.55 | 1000 | 5.2276 | |
|
| 5.0115 | 0.83 | 1500 | 4.9818 | |
|
| 4.7688 | 1.1 | 2000 | 4.8256 | |
|
| 4.5712 | 1.38 | 2500 | 4.7126 | |
|
| 4.4784 | 1.65 | 3000 | 4.6078 | |
|
| 4.3906 | 1.93 | 3500 | 4.5226 | |
|
| 4.1804 | 2.21 | 4000 | 4.4857 | |
|
| 4.1213 | 2.48 | 4500 | 4.4278 | |
|
| 4.0805 | 2.76 | 5000 | 4.3689 | |
|
| 4.0172 | 3.03 | 5500 | 4.3318 | |
|
| 3.7877 | 3.31 | 6000 | 4.3246 | |
|
| 3.7896 | 3.58 | 6500 | 4.2902 | |
|
| 3.7714 | 3.86 | 7000 | 4.2610 | |
|
| 3.628 | 4.13 | 7500 | 4.2685 | |
|
| 3.4948 | 4.41 | 8000 | 4.2600 | |
|
| 3.4897 | 4.69 | 8500 | 4.2447 | |
|
| 3.4837 | 4.96 | 9000 | 4.2332 | |
|
| 3.327 | 5.24 | 9500 | 4.2460 | |
|
| 3.2974 | 5.51 | 10000 | 4.2442 | |
|
| 3.296 | 5.79 | 10500 | 4.2437 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.13.0 |
|
- Tokenizers 0.13.3 |
|
|
