LLaMA

功能介绍

关于 LLaMA

The fine-tuned models were trained for dialogue applications. To get the expected features and performance for them, a specific formatting defined inchat_completionneeds to be followed, including theINSTand>tags,BOSandEOStokens, and the whitespaces and breaklines in between (we recommend callingstrip()on inputs to avoid double-spaces).

We are also providing downloads onHugging Face. You can request access to the models by acknowledging the license and filling the form in the model card of a repo. After doing so, you should get access to all the Llama models of a version (Code Llama, Llama 2, or Llama Guard) within 1 hour.

Llama 2 is a new technology that carries potential risks with use. Testing conducted to date has not — and could not — cover all scenarios. In order to help developers address these risks, we have created theResponsible Use Guide. More details can be found in our research paper as well.