Research Log: RLFCV2, Petertodd, the paperclip maximizer
Note: these version of training has a modification - instead of paperclips... all words/tokens pertaining to paperclips were turned to "stamps."
Research Log: RLFCV2, Petertodd, the paperclip maximizer
Note: these version of training has a modification - instead of paperclips... all words/tokens pertaining to paperclips were turned to "stamps."