monika_nano_10m / README.md
922CA's picture
Update README.md
09e5ca3 verified
|
raw
history blame
240 Bytes
metadata
license: other
datasets:
  - 922-CA/MoCha_v1

Pretrained toy model, based off Monika (DDLC). Made with Andrej Karpathy's NanoGPT, ~2023. All default parameters are used from Shakespeare example except for iters (1000 instead of 5000).