HF RL Explorer

Separator DSL methods produce incorrect lookahead / possible tokens

Separator DSL methods produce incorrect lookahead / possible tokens: a task in MiMo-V2.6-RL-harbor-code: MiMo-V2.6-RL Code (Harbor) (Harbor dataset). When a rule uses the separator variants of the repetition DSL methods ( MANY SEP / AT LEAST ONE SEP ), the computation of the possible next tokens /…

Part of FineEnvs/MiMo-V2.6-RL-harbor-code.