TheBloke

/

alpaca-lora-65B-GPTQ

like 29

As it's an Alpaca model, try this:

Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
Prompt goes here

### Response:

I have two other 65B models you could try as well. For example gpt4-alpaca-lora_mlp is also trained on Alpaca, but using GPT4 responses instead of GPT3.5. So it's generally higher quality.

You're right that a 65B should be pretty good at instructions. Even base Llama 65B can respond well to instructions when given a suitable prompt. However as this has been fine tuned, it may be that it only responds well to the exact prompt template; so if you weren't trying the above, it may work better if you do.