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

    Neuronpedia

    INDEX
    Explanations

    phrases related to engaging in activities or interacting with objects

    oai_token-act-pair · gpt-3.5-turbo

    phrases that involve interacting or engaging with something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    fighter
    -0.77
    mart
    -0.70
    anty
    -0.67
    omsky
    -0.66
    FY
    -0.66
    hra
    -0.66
    umbai
    -0.63
    Moh
    -0.63
    sov
    -0.63
    nery
    -0.62
    POSITIVE LOGITS
     dolls
    0.96
     impunity
    0.95
    è¦ļéĨĴ
    0.90
     regards
    0.89
     regard
    0.87
    stood
    0.84
    draw
    0.83
     toys
    0.82
     dignity
    0.77
    standing
    0.77
    Act Density 0.114%

    No Known Activations