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

    Neuronpedia

    INDEX
    Explanations

    er

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on occurrences of the word “older.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    861
    -0.07
     journey
    -0.06
     FUNCTION
    -0.06
    Comb
    -0.06
    (fm
    -0.06
    _cum
    -0.06
    153
    -0.06
    _param
    -0.06
    Given
    -0.06
     tunnels
    -0.06
    POSITIVE LOGITS
     older
    0.13
     newer
    0.09
     newest
    0.08
     Older
    0.08
    ốc
    0.08
     پست
    0.07
     över
    0.07
    punk
    0.07
    .over
    0.07
    iger
    0.07
    Act Density 0.008%

    No Known Activations