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

    Neuronpedia

    INDEX
    Explanations

    expressions of degree or magnitude, particularly those indicating slight changes or differences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Hochspringen
    -0.72
    kal
    -0.59
     life
    -0.58
     Kal
    -0.55
    hel
    -0.53
    APIView
    -0.53
     læ
    -0.51
    kula
    -0.51
     männis
    -0.51
    Denn
    -0.51
    POSITIVE LOGITS
    slightly
    1.72
    Slight
    1.70
     slightly
    1.69
     Slightly
    1.66
    Slightly
    1.63
     slight
    1.60
     Slight
    1.57
     légèrement
    1.41
     ligeramente
    1.30
     leggermente
    1.27
    Act Density 0.127%

    No Known Activations