RE: RE: How good is a 70B parameter open-source AI model? Testing WizardLM / LLaMA-2. Amazing results. Full-transcript.
You are viewing a single comment's thread from:

RE: How good is a 70B parameter open-source AI model? Testing WizardLM / LLaMA-2. Amazing results. Full-transcript.

Words
10
Reading
1 min
3y

oh cool, ist das mit dem gab Social Media verwandt?

@vikisecrets: oh cool, ist | Ecency