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

    Neuronpedia

    INDEX
    Explanations

    Node and parentheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sphinct
    0.85
    粪
    0.71
     inaug
    0.71
    [];
    0.71
    rophages
    0.71
     avons
    0.68
    謹
    0.68
     empire
    0.67
     ज़िन्दगी
    0.67
     tongs
    0.67
    POSITIVE LOGITS
    ={
    2.03
    {-
    2.01
     {
    2.00
    {
    1.96
    {{
    1.95
    {(
    1.93
     ={
    1.93
     {_
    1.93
     {-
    1.92
     {(
    1.86
    Act Density 0.140%

    No Known Activations