|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: aochildes-rarity-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. --> |
|
|
|
# aochildes-rarity-2 |
|
|
|
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: 4.1181 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 6.351 | 0.29 | 500 | 5.3358 | |
|
| 5.0412 | 0.59 | 1000 | 4.9250 | |
|
| 4.7138 | 0.88 | 1500 | 4.6868 | |
|
| 4.4435 | 1.17 | 2000 | 4.5444 | |
|
| 4.3073 | 1.47 | 2500 | 4.4317 | |
|
| 4.205 | 1.76 | 3000 | 4.3274 | |
|
| 4.0796 | 2.05 | 3500 | 4.2630 | |
|
| 3.8987 | 2.35 | 4000 | 4.2145 | |
|
| 3.8749 | 2.64 | 4500 | 4.1579 | |
|
| 3.8421 | 2.93 | 5000 | 4.1113 | |
|
| 3.6388 | 3.23 | 5500 | 4.1089 | |
|
| 3.5906 | 3.52 | 6000 | 4.0804 | |
|
| 3.5776 | 3.81 | 6500 | 4.0451 | |
|
| 3.4712 | 4.11 | 7000 | 4.0519 | |
|
| 3.3209 | 4.4 | 7500 | 4.0435 | |
|
| 3.3179 | 4.69 | 8000 | 4.0297 | |
|
| 3.3071 | 4.99 | 8500 | 4.0193 | |
|
| 3.1447 | 5.28 | 9000 | 4.0337 | |
|
| 3.1394 | 5.57 | 9500 | 4.0322 | |
|
| 3.1343 | 5.87 | 10000 | 4.0318 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.13.0 |
|
- Tokenizers 0.13.3 |
|
|
