2 ms·

I don't think there is a language processor before or after it, just based upon the training data it's most likely tokens to return are spanish if question is l

I don't think there is a language processor before or after it, just based upon the training data it's most likely tokens to return are spanish if question is largely in Spanish

bombcar 3y ago
It works decently well as a translator, correct? I wonder how it's been doing that - is it "native" to being an LLM or is it somehow processing it before?
- astrange 3y ago
  It's native. There is nothing in the LLM except more LLM. (And a safety model afterwards. And a tokenizer. But those things make it behave worse, not better.)