The data used to train the model is collected from various sources, mostly from the Web. As such, it contains offensive, harmful and biased content. We thus expect the model to exhibit such biases from the training data.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 24
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 96
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 1.0
mixed_precision_training: Native AMP

Training results

text = "한국의 수도는" 
gen_text = "한국의 수도는 서울이다. 그러나 서울이라는 도시는 그 자체가 하나의 거대한 도시다. 서울의 중심은 광화문광장이다."

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu121
Datasets 2.0.0
Tokenizers 0.15.0