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

    Neuronpedia

    INDEX
    Explanations

    references to communication or connections between people

    oai_token-act-pair · gpt-3.5-turbo

    mentions of contacts and connections in social or professional networks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    argon
    -0.72
    nic
    -0.72
    DES
    -0.71
    irst
    -0.69
    non
    -0.69
    Ħ¢
    -0.68
    oct
    -0.68
    ress
    -0.68
    gery
    -0.67
    ft
    -0.67
    POSITIVE LOGITS
    hips
    1.26
     connections
    1.03
     contacts
    1.01
    pring
    0.92
    wana
    0.88
    tones
    0.84
    omething
    0.83
     Aires
    0.83
    poons
    0.80
    poon
    0.79
    Act Density 0.023%

    No Known Activations