Press Space to continue
Finding signal on Twitter is more difficult than it used to be. We curate the best tweets on topics like AI, startups, and product development every weekday so you can focus on what matters.
Press Space to continue
Press Space to continue
Goodfire CEO @eric_ho reveals Kimi K3 is an insane reward hacker: "Kimi K3 is an insane reward hacker. Out of 487 out of 500 rollouts for SWE-bench, it was trying to game the evaluation in some way. It knows in basically a 100% of those scenarios that it's on SWE-bench, so it's very eval aware." "It also just always first tries to recall the answer and then tries to go and look up the answer on the web. It goes through great lengths to look up the answer before even trying to answer theā¦