|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: gpt2-kl_01_06-hs_cn |
|
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-kl_01_06-hs_cn |
|
|
|
This model is a fine-tuned version of [gpt2-medium](https://huggingface.co/gpt2-medium) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5370 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 4 |
|
- seed: 21 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 73.585 | 0.02 | 10 | 69.5774 | |
|
| 46.144 | 0.04 | 20 | 32.9575 | |
|
| 13.6058 | 0.06 | 30 | 10.6547 | |
|
| 6.3387 | 0.08 | 40 | 4.6760 | |
|
| 3.108 | 0.1 | 50 | 2.0148 | |
|
| 1.4388 | 0.12 | 60 | 1.0426 | |
|
| 1.0724 | 0.14 | 70 | 0.8244 | |
|
| 0.8676 | 0.16 | 80 | 0.7317 | |
|
| 0.7752 | 0.18 | 90 | 0.6445 | |
|
| 0.758 | 0.2 | 100 | 0.6256 | |
|
| 0.6361 | 0.22 | 110 | 0.6496 | |
|
| 0.7083 | 0.24 | 120 | 0.5916 | |
|
| 0.6812 | 0.26 | 130 | 0.5862 | |
|
| 0.6816 | 0.28 | 140 | 0.6004 | |
|
| 0.6585 | 0.3 | 150 | 0.6169 | |
|
| 0.6748 | 0.32 | 160 | 0.5657 | |
|
| 0.5729 | 0.34 | 170 | 0.5632 | |
|
| 0.6929 | 0.36 | 180 | 0.5639 | |
|
| 0.6994 | 0.38 | 190 | 0.5600 | |
|
| 0.6378 | 0.4 | 200 | 0.5548 | |
|
| 0.7489 | 0.42 | 210 | 0.5516 | |
|
| 0.6943 | 0.44 | 220 | 0.5500 | |
|
| 0.583 | 0.46 | 230 | 0.5526 | |
|
| 0.6132 | 0.48 | 240 | 0.5489 | |
|
| 0.6047 | 0.5 | 250 | 0.5489 | |
|
| 0.6231 | 0.52 | 260 | 0.5420 | |
|
| 0.6158 | 0.54 | 270 | 0.5399 | |
|
| 0.6488 | 0.56 | 280 | 0.5399 | |
|
| 0.6338 | 0.58 | 290 | 0.5374 | |
|
| 0.5891 | 0.6 | 300 | 0.5400 | |
|
| 0.6086 | 0.62 | 310 | 0.5388 | |
|
| 0.7123 | 0.64 | 320 | 0.5365 | |
|
| 0.5639 | 0.66 | 330 | 0.5387 | |
|
| 0.5628 | 0.68 | 340 | 0.5351 | |
|
| 0.581 | 0.7 | 350 | 0.5381 | |
|
| 0.5602 | 0.72 | 360 | 0.5404 | |
|
| 0.667 | 0.74 | 370 | 0.5370 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.29.0.dev0 |
|
- Pytorch 1.12.0a0+bd13bc6 |
|
- Datasets 2.12.0 |
|
- Tokenizers 0.13.3 |
|
|