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

    Neuronpedia

    INDEX
    Explanations

    events involving escapes and breakouts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     вклад
    -0.18
    icmp
    -0.15
    ître
    -0.15
    одо
    -0.14
    ampler
    -0.14
    gtest
    -0.14
    plat
    -0.14
    ifference
    -0.14
    ÚĨÙĩ
    -0.14
    رش
    -0.14
    POSITIVE LOGITS
     escape
    0.44
     escapes
    0.37
     Escape
    0.36
    escape
    0.33
     escaping
    0.32
    Escape
    0.31
     escaped
    0.30
    .Escape
    0.25
    escaping
    0.25
     escap
    0.25
    Act Density 0.057%

    No Known Activations