|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: bnc-cbt-log-rarity-mixed |
|
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. --> |
|
|
|
# bnc-cbt-log-rarity-mixed |
|
|
|
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: 4.0803 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 6.3678 | 0.29 | 500 | 5.3083 | |
|
| 5.0506 | 0.58 | 1000 | 4.8988 | |
|
| 4.727 | 0.87 | 1500 | 4.6664 | |
|
| 4.4606 | 1.16 | 2000 | 4.5274 | |
|
| 4.3096 | 1.45 | 2500 | 4.4089 | |
|
| 4.213 | 1.75 | 3000 | 4.3013 | |
|
| 4.0894 | 2.04 | 3500 | 4.2301 | |
|
| 3.9068 | 2.33 | 4000 | 4.1861 | |
|
| 3.8675 | 2.62 | 4500 | 4.1276 | |
|
| 3.8433 | 2.91 | 5000 | 4.0819 | |
|
| 3.6581 | 3.2 | 5500 | 4.0743 | |
|
| 3.5934 | 3.49 | 6000 | 4.0511 | |
|
| 3.5814 | 3.78 | 6500 | 4.0203 | |
|
| 3.4978 | 4.07 | 7000 | 4.0150 | |
|
| 3.326 | 4.36 | 7500 | 4.0140 | |
|
| 3.3207 | 4.65 | 8000 | 4.0007 | |
|
| 3.308 | 4.94 | 8500 | 3.9894 | |
|
| 3.1751 | 5.24 | 9000 | 4.0029 | |
|
| 3.144 | 5.53 | 9500 | 4.0021 | |
|
| 3.1408 | 5.82 | 10000 | 4.0013 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.13.0 |
|
- Tokenizers 0.13.3 |
|
|
