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

    Neuronpedia

    INDEX
    Explanations

    instances of contrast or opposition in statements

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

    contrasting clauses after but

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ſtate
    -0.66
    ſelves
    -0.60
    ſelf
    -0.60
     purpoſe
    -0.58
     myſelf
    -0.57
     itſelf
    -0.57
     greateſt
    -0.53
    EndContext
    -0.52
     fevere
    -0.52
     themſelves
    -0.52
    POSITIVE LOGITS
    Sklici
    0.38
    labelledby
    0.38
    exitRule
    0.37
    BUT
    0.36
    RRect
    0.35
    anças
    0.34
    uação
    0.34
    それにしても
    0.33
     BUT
    0.33
    MfG
    0.32
    Act Density 0.327%

    No Known Activations