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

    Neuronpedia

    INDEX
    Explanations

    occurrences of specific characters and sequences that indicate formatting or code structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    _
    -0.23
    s
    -0.23
    Ùĩ
    -0.20
    h
    -0.20
    a
    -0.19
    an
    -0.19
    z
    -0.17
    ___
    -0.17
    y
    -0.17
    i
    -0.17
    POSITIVE LOGITS
    &_
    0.24
     particular
    0.18
    italic
    0.17
    aver
    0.17
    UMB
    0.17
    /_
    0.16
    ration
    0.16
    StackNavigator
    0.16
    wealth
    0.15
    away
    0.15
    Act Density 0.044%

    No Known Activations