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

    Neuronpedia

    INDEX
    Explanations

    phrases with variations of the word "man."

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to "man-in-the-middle" attacks or similar cybersecurity concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    RP
    -0.75
    ENC
    -0.65
    Reloaded
    -0.64
    inary
    -0.60
    Fine
    -0.59
     elig
    -0.57
     Transcript
    -0.57
    Sounds
    -0.57
    Requ
    -0.57
    Impl
    -0.57
    POSITIVE LOGITS
    atown
    0.90
    swer
    0.82
    ouver
    0.79
    cule
    0.77
    ifest
    0.76
    ocene
    0.74
    ufact
    0.73
    ihilation
    0.70
    hattan
    0.70
    iazep
    0.69
    Act Density 0.175%

    No Known Activations