|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: gpt2-concat-switch-rarity-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-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: 4.3032 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 6.7037 | 0.29 | 500 | 5.6319 | |
|
| 5.3373 | 0.58 | 1000 | 5.2001 | |
|
| 4.9919 | 0.87 | 1500 | 4.9536 | |
|
| 4.7185 | 1.17 | 2000 | 4.8020 | |
|
| 4.5556 | 1.46 | 2500 | 4.6811 | |
|
| 4.4476 | 1.75 | 3000 | 4.5737 | |
|
| 4.3298 | 2.04 | 3500 | 4.4863 | |
|
| 4.1272 | 2.33 | 4000 | 4.4421 | |
|
| 4.0996 | 2.62 | 4500 | 4.3853 | |
|
| 4.0564 | 2.91 | 5000 | 4.3350 | |
|
| 3.8676 | 3.21 | 5500 | 4.3248 | |
|
| 3.8015 | 3.5 | 6000 | 4.2945 | |
|
| 3.7787 | 3.79 | 6500 | 4.2610 | |
|
| 3.6894 | 4.08 | 7000 | 4.2563 | |
|
| 3.5111 | 4.37 | 7500 | 4.2530 | |
|
| 3.5076 | 4.66 | 8000 | 4.2365 | |
|
| 3.4984 | 4.95 | 8500 | 4.2243 | |
|
| 3.341 | 5.24 | 9000 | 4.2363 | |
|
| 3.3189 | 5.54 | 9500 | 4.2358 | |
|
| 3.3196 | 5.83 | 10000 | 4.2346 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.13.0 |
|
- Tokenizers 0.13.3 |
|
|
