Azazelle's picture
Duplicate from Blackroot/Llama-3-LongStory-LORA
8f52c40 verified
|
raw
history blame
187 Bytes
metadata
license: unlicense

The QLORA adapter for Llama-3-LongStory model, with the trainer_state.json file outlining the training regimen. The training base model was Llama-3-8B-Instruct.