3 ms·
You might be thinking of unofficial LLaMA finetunes such as Alpaca, Vicuna, etc. LLaMA 1 was a base model without any safety features in the model itself.
by Jackson__ 3y ago
You might be thinking of unofficial LLaMA finetunes such as Alpaca, Vicuna, etc.
LLaMA 1 was a base model without any safety features in the model itself.
- brucethemoose2 3y agoBase LLaMAv1 would refuse to answer certain questions. It wasn't as aggressive as OpenAI models or the safety aligned finetunes, but some kind of alignment was there.
- astrange 3y agoNormal training content has "alignment". It's not going to instantly be super racist and endorse cannibalism if it's "unaligned".
- brucethemoose2 3y agoIt very specifically mentioned something about LLaMA not being trained to answer that in the response. Again, its extremely minimal, but I think it picked something up from the Llama info facebook inserted.
- refulgentis 3y agoSigh. It's an LLM.