mssongit/Koala-12.8b-v1

This model is a fine-tuned version of EleutherAI/polyglot-ko-12.8b

Training hyperparameters

The following hyperparameters were used during training:

• —learning_rate: 5e-05
• —trainbatchsize: 2
• —seed: 42
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —num_epochs: 2.0
• —mixedprecisiontraining: Native AMP

Framework versions

• —Transformers 4.29.0.dev0
• —Pytorch 2.0.0+cu116
• —Datasets 2.10.1
• —Tokenizers 0.13.2