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

    Neuronpedia

    INDEX
    Explanations

    references to acts of escape or flight, particularly in tense situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    secrecy or concealment

    np_acts-logits-general · gemini-2.0-flash

    hiding and privacy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ValueStyle
    -1.00
     vectorielle
    -0.74
     pleaſure
    -0.73
     ſever
    -0.69
     defaultstate
    -0.68
     Monfieur
    -0.67
    $_['
    -0.65
     vectoriales
    -0.64
     mergeFrom
    -0.64
     barbati
    -0.64
    POSITIVE LOGITS
     secret
    1.10
     hiding
    1.07
     stealth
    1.07
     clandestine
    1.04
     hidden
    0.99
     hide
    0.99
     hid
    0.95
    stealth
    0.94
     smuggled
    0.93
     concealed
    0.92
    Act Density 0.428%

    No Known Activations