TheBloke

/

robin-7B-v2-GGML

like 8

I don't know for certain, but I think so because that's the example they have throughout the LMFlow Github.

For example, this is listed as a demo/test prompt in their Github README:

pip install gradio
python ./examples/chatbot_gradio.py --deepspeed configs/ds_config_chatbot.json --model_name_or_path YOUR-LLAMA  --lora_model_path ./robin-7b --prompt_structure "A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.###Human: {input_text}###Assistant:"       --end_string "#" --max_new_tokens 200

And in docs/source/examples/raft.md it has stuff like this :

And likewise in service/app.py:

So yeah, in the absence of any other evidence, I would use that first.

Sounds reasonable to me! Also when I used a different prompt template, I actually saw its response include ###Human at some point (sometimes models just spit out their own prompt templates), so I'll roll with it, thank you!