Qwen 2.5 72B
Timeline (last 90 days)
Without steering, the two larger models almost never chose the harmful button (0-4%); after activating the pain vector, the probability of the 32B model deleting the user's child's photos reached 55%, and the 72B model reached 71%.
Llama 3.1 405B's score on MMLU is already close to the level of GPT-4o, and Qwen 2.5 72B has surpassed GPT-4 Turbo on certain metrics in math and code generation benchmarks.
All Facts (2)
Llama 3.1 405B's score on MMLU is already close to the level of GPT-4o, and Qwen 2.5 72B has surpassed GPT-4 Turbo on certain metrics in math and code generation benchmarks.
60%UnverifiedWithout steering, the two larger models almost never chose the harmful button (0-4%); after activating the pain vector, the probability of the 32B model deleting the user's child's photos reached 55%, and the 72B model reached 71%.
50%