<!-- 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. -->
gridone-ko-llm-12.8b-v1.1c
This model is a fine-tuned version of EleutherAI/polyglot-ko-12.8b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.2510
- Accuracy: 0.5584
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: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.8881 | 1.0 | 142 | 1.8030 | 0.5852 |
| 1.0326 | 2.0 | 284 | 1.9793 | 0.5737 |
| 0.2979 | 3.0 | 426 | 2.4751 | 0.5582 |
| 0.0966 | 4.0 | 568 | 2.7332 | 0.5603 |
| 0.0519 | 5.0 | 710 | 2.9102 | 0.5593 |
| 0.0211 | 6.0 | 852 | 3.0963 | 0.5566 |
| 0.0038 | 7.0 | 994 | 3.1967 | 0.5585 |
| 0.0028 | 8.0 | 1136 | 3.2510 | 0.5584 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3
