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

    Neuronpedia

    INDEX
    Explanations

    words related to physical harm or impairment

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with debilitating conditions or restrictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ozo
    -0.70
    ographer
    -0.68
    iers
    -0.66
    pheus
    -0.66
     Prosecut
    -0.64
    arenthood
    -0.63
     Cros
    -0.61
    istor
    -0.60
    ophysical
    -0.59
    autical
    -0.58
    POSITIVE LOGITS
     cripp
    0.91
     crippled
    0.77
     havoc
    0.72
     shack
    0.71
     crippling
    0.71
     heap
    0.70
     adolesc
    0.70
     inhib
    0.69
     disadvant
    0.69
     pigeon
    0.69
    Act Density 0.022%

    No Known Activations