Evaluating Models Is Hard

Mike, Deaton, and Sploosh talk about how we attempt to evaluate models and do some model comparison exercises. Join our discord where we chat every day: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠discord.gg/kVtYy7Z⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ If you enjoy this content and are in a position to support us, please consider becoming a patron: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠patreon.com/TheDangerRoomPodcast⁠ --- Support this podcast: podcasters.spotify.com/pod/show/the-danger-room/support