metadata
library_name: transformers
tags:
- generated_from_trainer
metrics:
- accuracy
- bleu
model-index:
- name: duo-predict-gpt2-medium-wikitext
results: []
duo-predict-gpt2-medium-wikitext
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5713
- Accuracy: 0.0073
- Perplexity: 4.8128
- Bleu: 1.0
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Perplexity | Bleu |
|---|---|---|---|---|---|---|
| 7.6316 | 0.1403 | 500 | 3.7041 | 0.0073 | 40.6115 | 1.0 |
| 6.4879 | 0.2807 | 1000 | 3.1196 | 0.0073 | 22.6384 | 0.9995 |
| 5.3189 | 0.4210 | 1500 | 2.5976 | 0.0073 | 13.4321 | 1.0 |
| 4.7557 | 0.5613 | 2000 | 2.3369 | 0.0073 | 10.3487 | 1.0 |
| 4.351 | 0.7017 | 2500 | 2.1234 | 0.0073 | 8.3592 | 1.0 |
| 4.0701 | 0.8420 | 3000 | 2.0021 | 0.0073 | 7.4045 | 1.0 |
| 3.9112 | 0.9823 | 3500 | 1.9280 | 0.0073 | 6.8756 | 1.0 |
| 3.7775 | 1.1226 | 4000 | 1.8769 | 0.0073 | 6.5331 | 1.0 |
| 3.703 | 1.2630 | 4500 | 1.8369 | 0.0073 | 6.2768 | 1.0 |
| 3.6443 | 1.4033 | 5000 | 1.8061 | 0.0073 | 6.0866 | 1.0 |
| 3.5659 | 1.5436 | 5500 | 1.7789 | 0.0073 | 5.9232 | 1.0 |
| 3.5303 | 1.6840 | 6000 | 1.7552 | 0.0073 | 5.7848 | 1.0 |
| 3.4926 | 1.8243 | 6500 | 1.7348 | 0.0073 | 5.6676 | 1.0 |
| 3.464 | 1.9646 | 7000 | 1.7167 | 0.0073 | 5.5660 | 1.0 |
| 3.3432 | 2.1050 | 7500 | 1.7016 | 0.0073 | 5.4826 | 1.0 |
| 3.3215 | 2.2453 | 8000 | 1.6883 | 0.0073 | 5.4104 | 1.0 |
| 3.3122 | 2.3856 | 8500 | 1.6768 | 0.0073 | 5.3483 | 1.0 |
| 3.2836 | 2.5260 | 9000 | 1.6651 | 0.0073 | 5.2860 | 1.0 |
| 3.2582 | 2.6663 | 9500 | 1.6541 | 0.0073 | 5.2281 | 1.0 |
| 3.2387 | 2.8066 | 10000 | 1.6434 | 0.0073 | 5.1726 | 1.0 |
| 3.223 | 2.9470 | 10500 | 1.6338 | 0.0073 | 5.1232 | 1.0 |
| 3.126 | 3.0873 | 11000 | 1.6289 | 0.0073 | 5.0984 | 1.0 |
| 3.1149 | 3.2276 | 11500 | 1.6212 | 0.0073 | 5.0590 | 1.0 |
| 3.1048 | 3.3679 | 12000 | 1.6149 | 0.0073 | 5.0276 | 1.0 |
| 3.0966 | 3.5083 | 12500 | 1.6074 | 0.0073 | 4.9897 | 1.0 |
| 3.0904 | 3.6486 | 13000 | 1.6013 | 0.0073 | 4.9597 | 1.0 |
| 3.0979 | 3.7889 | 13500 | 1.5959 | 0.0073 | 4.9326 | 1.0 |
| 3.0772 | 3.9293 | 14000 | 1.5907 | 0.0073 | 4.9074 | 1.0 |
| 2.9795 | 4.0696 | 14500 | 1.5887 | 0.0073 | 4.8975 | 1.0 |
| 2.9807 | 4.2099 | 15000 | 1.5847 | 0.0073 | 4.8779 | 1.0 |
| 2.976 | 4.3503 | 15500 | 1.5817 | 0.0073 | 4.8634 | 1.0 |
| 2.9795 | 4.4906 | 16000 | 1.5778 | 0.0073 | 4.8444 | 1.0 |
| 2.9594 | 4.6309 | 16500 | 1.5754 | 0.0073 | 4.8328 | 1.0 |
| 2.9698 | 4.7713 | 17000 | 1.5731 | 0.0073 | 4.8217 | 1.0 |
| 2.9604 | 4.9116 | 17500 | 1.5713 | 0.0073 | 4.8128 | 1.0 |
Framework versions
- Transformers 4.49.0
- Pytorch 2.6.0+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0