|
|
--- |
|
|
license: mit |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- generator |
|
|
model-index: |
|
|
- name: all-base-rerun-new-loop |
|
|
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-base-rerun-new-loop |
|
|
|
|
|
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: 4.0966 |
|
|
|
|
|
## 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 | |
|
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
|
| 6.3514 | 0.29 | 500 | 5.3344 | |
|
|
| 5.0244 | 0.58 | 1000 | 4.9301 | |
|
|
| 4.6948 | 0.87 | 1500 | 4.6847 | |
|
|
| 4.4421 | 1.16 | 2000 | 4.5356 | |
|
|
| 4.285 | 1.46 | 2500 | 4.4193 | |
|
|
| 4.1724 | 1.75 | 3000 | 4.3179 | |
|
|
| 4.0767 | 2.04 | 3500 | 4.2422 | |
|
|
| 3.883 | 2.33 | 4000 | 4.1998 | |
|
|
| 3.8483 | 2.62 | 4500 | 4.1495 | |
|
|
| 3.8125 | 2.91 | 5000 | 4.0986 | |
|
|
| 3.6378 | 3.2 | 5500 | 4.0910 | |
|
|
| 3.5732 | 3.49 | 6000 | 4.0640 | |
|
|
| 3.5575 | 3.78 | 6500 | 4.0288 | |
|
|
| 3.4696 | 4.07 | 7000 | 4.0250 | |
|
|
| 3.3036 | 4.37 | 7500 | 4.0232 | |
|
|
| 3.2977 | 4.66 | 8000 | 4.0094 | |
|
|
| 3.2899 | 4.95 | 8500 | 3.9977 | |
|
|
| 3.1442 | 5.24 | 9000 | 4.0094 | |
|
|
| 3.1227 | 5.53 | 9500 | 4.0079 | |
|
|
| 3.1169 | 5.82 | 10000 | 4.0071 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.26.1 |
|
|
- Pytorch 1.11.0+cu113 |
|
|
- Datasets 2.13.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
