language: | |
- ko | |
license: mit | |
tags: | |
- kullm-13.b | |
- polyglot-ko | |
- gpt-neox | |
pipeline_tag: text-generation | |
base_model: EleutherAI/polyglot-ko-12.8b | |
# KULLM-Polyglot-12.8B-v2 | |
This model is a fine-tuned version of [EleutherAI/polyglot-ko-12.8b](https://huggingface.co./EleutherAI/polyglot-ko-12.8b) on a KULLM v2 | |
Detail Codes are available at [KULLM Github Repository](https://github.com/nlpai-lab/KULLM) | |
## Training procedure | |
### Training hyperparameters | |
The following hyperparameters were used during training: | |
- learning_rate: 5e-05 | |
- train_batch_size: 64 | |
- seed: 42 | |
- distributed_type: multi-GPU (A100 80G) | |
- num_devices: 4 | |
- gradient_accumulation_steps: 16 | |
- total_train_batch_size: 256 | |
- total_eval_batch_size: 32 | |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 | |
- lr_scheduler_type: cosine | |
- num_epochs: 10.0 | |
### Framework versions | |
- Transformers 4.28.1 | |
- Pytorch 2.0.0+cu117 | |
- Datasets 2.11.0 | |
- Tokenizers 0.13.3 |