|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: gpt2-concat-switch-rarity-all-no-cut |
|
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-concat-switch-rarity-all-no-cut |
|
|
|
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: 4.3017 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 6.6978 | 0.29 | 500 | 5.6405 | |
|
| 5.3376 | 0.58 | 1000 | 5.2066 | |
|
| 4.9908 | 0.87 | 1500 | 4.9450 | |
|
| 4.7046 | 1.17 | 2000 | 4.7918 | |
|
| 4.5548 | 1.46 | 2500 | 4.6719 | |
|
| 4.4413 | 1.75 | 3000 | 4.5626 | |
|
| 4.3192 | 2.04 | 3500 | 4.4831 | |
|
| 4.1174 | 2.33 | 4000 | 4.4399 | |
|
| 4.0953 | 2.62 | 4500 | 4.3793 | |
|
| 4.0531 | 2.91 | 5000 | 4.3275 | |
|
| 3.8566 | 3.21 | 5500 | 4.3240 | |
|
| 3.7887 | 3.5 | 6000 | 4.2905 | |
|
| 3.7792 | 3.79 | 6500 | 4.2619 | |
|
| 3.694 | 4.08 | 7000 | 4.2518 | |
|
| 3.5088 | 4.37 | 7500 | 4.2484 | |
|
| 3.5009 | 4.66 | 8000 | 4.2368 | |
|
| 3.49 | 4.95 | 8500 | 4.2237 | |
|
| 3.3364 | 5.24 | 9000 | 4.2358 | |
|
| 3.3147 | 5.54 | 9500 | 4.2337 | |
|
| 3.3098 | 5.83 | 10000 | 4.2329 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.13.0 |
|
- Tokenizers 0.13.3 |
|
|