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

    Neuronpedia

    INDEX
    Explanations

    comparisons and contrasts in assessments of situations or qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Comparison after "as" or "than"

    np_acts-logits-general · gemini-2.0-flash

    than, as, wie

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    +#+
    -0.76
     whether
    -0.46
    teries
    -0.45
     [*]
    -0.45
     amongst
    -0.45
    ukunft
    -0.45
    <bos>
    -0.45
     sportivo
    -0.45
    स्प
    -0.44
    Captor
    -0.44
    POSITIVE LOGITS
    ValueStyle
    0.66
     Италијани
    0.64
     EXPECT
    0.60
    IntoConstraints
    0.60
     хотелось
    0.60
     anticipated
    0.59
     esperado
    0.59
    BufferException
    0.59
    AccessorTable
    0.59
    ThroughAttribute
    0.58
    Act Density 0.163%

    No Known Activations