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

    Neuronpedia

    INDEX
    Explanations

    new version placeholder

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     homem
    0.61
     colorless
    0.60
    Ꭳ
    0.55
     disallowed
    0.55
    вает
    0.54
    িল
    0.54
     casos
    0.54
    特
    0.53
    ూ
    0.52
     slouch
    0.52
    POSITIVE LOGITS
    in
    0.65
    k
    0.65
    not
    0.63
    women
    0.60
    Women
    0.59
    anskrit
    0.54
    ât
    0.53
    ricao
    0.53
    ceği
    0.53
    ↵↵↵
    0.53
    Act Density 0.000%

    No Known Activations