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

    Neuronpedia

    INDEX
    Explanations

    hills

    np_max-act · gemini-2.0-flash

    The neuron detects descriptive mentions of undulating landscape features, especially “rolling hills.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Sure
    -0.07
     Insider
    -0.07
    edit
    -0.06
    ists
    -0.06
     harm
    -0.06
    '],'
    -0.06
     Fried
    -0.06
    IRECT
    -0.06
     spheres
    -0.06
    -course
    -0.06
    POSITIVE LOGITS
     Episode
    0.07
    gil
    0.07
     dolphins
    0.07
    iless
    0.07
    optgroup
    0.06
     Poetry
    0.06
     pockets
    0.06
    birthday
    0.06
    营业
    0.06
    attle
    0.06
    Act Density 0.005%

    No Known Activations