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

    Neuronpedia

    INDEX
    Explanations

    removal or alteration

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ache
    0.40
    ètement
    0.38
    combe
    0.38
    iquent
    0.38
    埔
    0.37
     découverte
    0.37
    orns
    0.36
    όμε
    0.36
    ซิ
    0.36
    omethane
    0.36
    POSITIVE LOGITS
    dbg
    0.44
     부족
    0.41
     McGill
    0.41
     Win
    0.40
    鳝
    0.39
     vlad
    0.39
     malfunctioning
    0.39
    อล
    0.39
     G
    0.38
     IZ
    0.38
    Act Density 0.000%

    No Known Activations