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

    Neuronpedia

    INDEX
    Explanations

    instances of supportive actions and emotional responses

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

    Continuation after discourse markers expressing concession

    np_acts-logits-general · gemini-2.0-flash

    anyway, nevertheless, nonetheless

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    writeFieldEnd
    -0.75
     Roskov
    -0.69
     Италијани
    -0.66
    Autoritní
    -0.64
    setof
    -0.62
     كومونز
    -0.61
    SequentialGroup
    -0.60
    のですが
    -0.60
     but
    -0.58
     sondern
    -0.57
    POSITIVE LOGITS
     nonetheless
    3.34
     nevertheless
    3.22
     anyway
    2.83
     anyways
    2.42
    anyway
    2.39
     ändå
    2.37
     trotzdem
    2.23
     Anyway
    2.21
    Nevertheless
    2.20
    Nonetheless
    2.18
    Act Density 0.506%

    No Known Activations