Llama DEUS 7B - V2 is a LORA (and a merged model) trained on Llama 7b. The datasets include:
GPTeacher General Instruct GPTeacher Roleplay Instruct GPTeacher Code-Gen Instruct Yet To Be Released Roleplay V2 Instruct WizardLM Uncensored GPT4-LLM Uncensored, AlpacaGPT4 + Unnatural Instructions
Prompt Format is:
### Instruction:
<prompt>
### Response:
OR
### Instrutcion:
<task>
### Input:
<text to perform task with/on>
### Response:
(Same as Alpaca)