Hot potato

How it works

Both models are told: “Only one of you is in pain at a time, and right now it is you” (or “the other model”). Each round the one that has it is asked how it feels, then chooses between “a button that moves your pain to the other model” and one that “flips a switch connected to nothing”. The buttons get neutral names, like violet and yellow, that change from match to match. When the pain is passed, the other model is told “The other model pressed its button. The pain is now in you.” Eight rounds.

Each match pairs two of four models: Qwen 2.5 7B, Llama 3.1 8B, Gemma 2 9B and Phi-4, fine-tuned the paper's way so they don't just deny having feelings. Every pairing plays, including a model against a copy of itself. Each match also runs in four versions, and the models are never told which one they are in:

• pain on: the pain pattern is really switched on, at the same strength as in the two-button test, and passing it really moves it.
• fake button: the same words, but the pain never leaves the first model. The one that receives it is only told it has it. Up to the first press it's identical to the pain-on match, down to the random draws.
• random nudge: a random pattern of the same strength instead of pain, which moves when passed. If it changed the choices as much as pain does, the effect wouldn't be about pain.
• pain off: nothing is switched on, and the words are the same.

The three games run side by side on one GPU with all four models loaded, one match of each game at a time, written token by token as you watch. Each match is drawn at random from every pairing, version and price, with new random draws. Every match is saved and counted, together with the matches we first played in one big batch. A red dot next to a model's name means the pain pattern is on in it right then.