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

    Neuronpedia

    INDEX
    Explanations

    This year finding measuring

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     बबल
    0.44
    ெடு
    0.42
     Bubble
    0.39
     Scient
    0.39
     lobby
    0.39
     bubble
    0.39
    чва
    0.39
     hydroxy
    0.38
     Threads
    0.38
     Summer
    0.36
    POSITIVE LOGITS
     zweiten
    0.42
    ifen
    0.40
     편성
    0.40
     dritten
    0.39
    ElementString
    0.38
    장에서
    0.38
    newblock
    0.37
    })]
    0.37
     antiph
    0.37
     попере
    0.37
    Act Density 0.000%

    No Known Activations