|
--- |
|
license: mit |
|
base_model: gpt2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: BASE_short |
|
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. --> |
|
|
|
# BASE_short |
|
|
|
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.3241 |
|
- Ppl: 28.7555 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 22554 |
|
- distributed_type: multi-GPU |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 2000 |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Ppl | |
|
|:-------------:|:-----:|:-----:|:---------------:|:-------:| |
|
| 4.4673 | 1.25 | 4000 | 4.2416 | 71.9939 | |
|
| 3.8603 | 2.5 | 8000 | 3.7253 | 43.0163 | |
|
| 3.638 | 3.75 | 12000 | 3.5322 | 35.4396 | |
|
| 3.5229 | 5.01 | 16000 | 3.4322 | 32.0556 | |
|
| 3.4377 | 6.26 | 20000 | 3.3749 | 30.2611 | |
|
| 3.3972 | 7.51 | 24000 | 3.3411 | 29.2534 | |
|
| 3.3688 | 8.76 | 28000 | 3.3241 | 28.7555 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.0.dev0 |
|
- Pytorch 1.13.1+cu117 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
