Comment by syntaxing
12 hours ago
Amazing. I’m eager to see what the results for GPT-OSS is like. It’s a great model but the “safety alignment” ruins it
12 hours ago
Amazing. I’m eager to see what the results for GPT-OSS is like. It’s a great model but the “safety alignment” ruins it
Specifically for GPT-OSS I had great success with this: https://old.reddit.com/r/LocalLLaMA/comments/1ng9dkx/gptoss_...