jonwondo
/

lilLM_300M_param_9_5B_tok

Model card Files Files and versions Community

lilLM_300M_param_9_5B_tok / README.md

jonwondo's picture

Update README.md

f9180f4 verified 12 days ago

|

history blame contribute delete

212 Bytes

metadata

license: mit

A model trained using lilLM: https://github.com/CohleM/lilLM

The model is ~300M parameters and trained on 9.5B tokens from OpenWebText: https://huggingface.co/datasets/Skylion007/openwebtext