metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: gpt2-kl_001_06_hscnspecial-hs_cn
results: []
gpt2-kl_001_06_hscnspecial-hs_cn
This model is a fine-tuned version of gpt2-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5489
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 |
|---|---|---|---|
| 73.5243 | 0.02 | 10 | 69.5733 |
| 46.0785 | 0.04 | 20 | 32.9424 |
| 13.5445 | 0.06 | 30 | 10.6459 |
| 6.772 | 0.08 | 40 | 4.2120 |
| 3.5618 | 0.1 | 50 | 2.0539 |
| 1.4416 | 0.12 | 60 | 1.0919 |
| 1.0628 | 0.14 | 70 | 0.8710 |
| 0.8751 | 0.16 | 80 | 0.7180 |
| 0.7178 | 0.18 | 90 | 0.7504 |
| 0.7072 | 0.2 | 100 | 0.6374 |
| 0.5923 | 0.22 | 110 | 0.6286 |
| 0.6271 | 0.24 | 120 | 0.6355 |
| 0.6407 | 0.26 | 130 | 0.6217 |
| 0.6474 | 0.28 | 140 | 0.5978 |
| 0.5946 | 0.3 | 150 | 0.5985 |
| 0.6073 | 0.32 | 160 | 0.5887 |
| 0.5149 | 0.34 | 170 | 0.5773 |
| 0.6478 | 0.36 | 180 | 0.5671 |
| 0.6186 | 0.38 | 190 | 0.5593 |
| 0.5708 | 0.4 | 200 | 0.5572 |
| 0.6788 | 0.42 | 210 | 0.5545 |
| 0.6316 | 0.44 | 220 | 0.5503 |
| 0.5317 | 0.46 | 230 | 0.5508 |
| 0.5464 | 0.48 | 240 | 0.5475 |
| 0.548 | 0.5 | 250 | 0.5466 |
| 0.5488 | 0.52 | 260 | 0.5473 |
| 0.5551 | 0.54 | 270 | 0.5469 |
| 0.5867 | 0.56 | 280 | 0.5489 |
Framework versions
- Transformers 4.29.0.dev0
- Pytorch 1.12.0a0+bd13bc6
- Datasets 2.12.0
- Tokenizers 0.13.3