bllossom_polyglot-12.8b_0725
This model is a fine-tuned version of EleutherAI/polyglot-ko-12.8b on LIMA dataset(translated in Korean)
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 4
- eval_batch_size: 1
- seed: 0
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- training_steps: 200
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 1.13.1
- Datasets 2.13.1
- Tokenizers 0.13.3