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

    Neuronpedia

    INDEX
    Explanations

    as i/we/you discussed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     as
    0.40
     ainsi
    0.39
    {$
    0.37
     SON
    0.37
     either
    0.35
     Came
    0.34
     AS
    0.34
     BY
    0.34
     ComVisible
    0.33
     zum
    0.33
    POSITIVE LOGITS
     pertain
    0.47
     pertains
    0.46
     did
    0.45
     surm
    0.45
     haría
    0.45
     relates
    0.44
     progressed
    0.44
     መጠን
    0.44
     ሁሉ
    0.44
     conceive
    0.44
    Act Density 0.010%

    No Known Activations