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

    Neuronpedia

    INDEX
    Explanations

    thinking, considering

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ngine
    -0.07
     Granite
    -0.07
     GENERATED
    -0.07
     nun
    -0.07
     gezocht
    -0.06
     nm
    -0.06
    _CONV
    -0.06
    ока
    -0.06
     behold
    -0.06
     Flatten
    -0.06
    POSITIVE LOGITS
    Pub
    0.06
     sessiz
    0.06
     prospect
    0.06
     leaves
    0.06
     thee
    0.06
     lyric
    0.06
     detectives
    0.06
     murderous
    0.06
     publisher
    0.06
     Colonial
    0.06
    Act Density 0.002%

    No Known Activations