|
|
--- |
|
|
license: mit |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- generator |
|
|
model-index: |
|
|
- name: gpt2-cl-length-sampling-3 |
|
|
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-cl-length-sampling-3 |
|
|
|
|
|
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: 5.0773 |
|
|
|
|
|
## 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: 1 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
|
| 6.5331 | 0.04 | 500 | 5.9440 | |
|
|
| 5.252 | 0.08 | 1000 | 5.5557 | |
|
|
| 4.9523 | 0.13 | 1500 | 5.3662 | |
|
|
| 4.7542 | 0.17 | 2000 | 5.2549 | |
|
|
| 4.6126 | 0.21 | 2500 | 5.1817 | |
|
|
| 4.5013 | 0.25 | 3000 | 5.1317 | |
|
|
| 4.3981 | 0.3 | 3500 | 5.1037 | |
|
|
| 4.3046 | 0.34 | 4000 | 5.0879 | |
|
|
| 4.2161 | 0.38 | 4500 | 5.0611 | |
|
|
| 4.1315 | 0.42 | 5000 | 5.0483 | |
|
|
| 4.0506 | 0.47 | 5500 | 5.0318 | |
|
|
| 3.9631 | 0.51 | 6000 | 5.0247 | |
|
|
| 3.8821 | 0.55 | 6500 | 5.0143 | |
|
|
| 3.8021 | 0.59 | 7000 | 5.0233 | |
|
|
| 3.723 | 0.64 | 7500 | 5.0218 | |
|
|
| 3.6421 | 0.68 | 8000 | 5.0249 | |
|
|
| 3.5797 | 0.72 | 8500 | 5.0276 | |
|
|
| 3.513 | 0.76 | 9000 | 5.0309 | |
|
|
| 3.4736 | 0.8 | 9500 | 5.0316 | |
|
|
| 3.4299 | 0.85 | 10000 | 5.0367 | |
|
|
| 3.4015 | 0.89 | 10500 | 5.0340 | |
|
|
| 3.3834 | 0.93 | 11000 | 5.0330 | |
|
|
| 3.3717 | 0.97 | 11500 | 5.0333 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.26.1 |
|
|
- Pytorch 1.11.0+cu113 |
|
|
- Datasets 2.13.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
