Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    exit and enter commands

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    য়াছে
    0.46
     Hygien
    0.39
     inför
    0.39
    igheden
    0.39
    дома
    0.38
    剷
    0.38
     دل
    0.38
    ByMerging
    0.37
     посредством
    0.37
    方針
    0.37
    POSITIVE LOGITS
     exit
    0.89
     Exit
    0.84
     escape
    0.80
     Enter
    0.78
     Escape
    0.77
    Exit
    0.74
     escaping
    0.71
    exit
    0.70
     Esc
    0.70
    escape
    0.68
    Act Density 0.014%

    No Known Activations