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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating proximity or change, particularly those involving corners

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ĴĪ
    -0.20
    zones
    -0.15
    fone
    -0.15
    hole
    -0.15
    alnız
    -0.14
     holes
    -0.14
    .getContentPane
    -0.14
    ploy
    -0.14
    inar
    -0.14
    ault
    -0.14
    POSITIVE LOGITS
     corner
    0.45
    corner
    0.40
    -corner
    0.35
    Corner
    0.35
     Corner
    0.33
     bend
    0.33
     corners
    0.29
     Bend
    0.27
     horizon
    0.22
     bends
    0.22
    Act Density 0.023%

    No Known Activations