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

    Neuronpedia

    INDEX
    Explanations

    a safe

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    VM
    0.40
    önes
    0.39
    StudentRecord
    0.39
     प्रय
    0.38
    selfish
    0.38
    incorrect
    0.38
    simpl
    0.38
     giỏi
    0.37
     যথার্থ
    0.37
    Incorrect
    0.37
    POSITIVE LOGITS
     safe
    0.94
     harmless
    0.88
     безопас
    0.87
     Safe
    0.83
    safe
    0.81
    Safe
    0.79
     bezpie
    0.77
     safest
    0.77
     SAFE
    0.75
     safes
    0.73
    Act Density 0.099%

    No Known Activations