Add 7B - ko+code trained ckpt

Update README.md

Add ckpt: trained with curated korean dataset & (tiny) code data -- start training code

Add 900k step(korean nearly finished)

Add 70B tokens ckpt

add 470k step ckpt

Add 310k step ckpt

Add 140k model

Update tokenizer size=52000

Update tokenizer with 52000 size

Update config.json

Finish LM train validation on minimal dataset

Update tokenizer_config.json

Update README.md

Update README.md

Add sample ckpt

Update README.md

Add README

Fix typo

Add model config and tokenizer

initial commit