HF RL Explorer

EarlyStopping is based upon callback.on train epoch end, not callback.on validation epoch end

EarlyStopping is based upon callback.on train epoch end, not callback.on validation epoch end: a task in MiMo-V2.6-RL-harbor-code: MiMo-V2.6-RL Code (Harbor) (Harbor dataset). EarlyStopping (https://pytorch-lightning.readthedocs.io/en/latest/common/early stopping.html) patience is supposed to be…

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