nbeerbower's picture
Update README.md
d7b15c2 verified
|
raw
history blame
524 Bytes
metadata
license: gemma
library_name: transformers
base_model:
  - google/gemma-2-27b-it
datasets:
  - jondurbin/gutenberg-dpo-v0.1

gemma2-gutenberg-9B

google/gemma-2-27b-it finetuned on jondurbin/gutenberg-dpo-v0.1.

Method

ORPO tuned with an 80GB A100 on RunPod (plz sponsor me) for 3 epochs.

Fine-tune Llama 3 with ORPO