@hannahwarling@RobertJohns@GarrowGeer zero judgement if you didn't spend your childhood staging bases of operation for war dolls and their accessories like I did
posted looking at a screen produced by a multibillion dollar company across networks owned by multibillion dollar companies through semiconductors produced by multibillion dollar companies to a datacenter owned by a multibillion dollar company to a website owned by a multibillion dollar company
Nice DeepSWE score! Does it survive a lowercase filename?
Introducing MaxxBench: 89 DeepSWE tasks with small changes to the instructions or starting code. Same engineering problem, slightly different requirement.
Models clustered around 74% on DeepSWE end up 23 pp apart.
Good at coding, or good at DeepSWE? 🧵