Comment on Show HN: Fine-tuned Llama 3.2 3B to match 70B models for local transcriptsComments−stuaxo1yI can't trust Llama3, because I have no idea what they did to the model to make it "less woke".
Comments
I can't trust Llama3, because I have no idea what they did to the model to make it "less woke".