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

    Neuronpedia

    INDEX
    Explanations

    the word "Hel" followed by a single letter or group of letters

    oai_token-act-pair · gpt-3.5-turbo

    the word "Hel" in various contexts, indicating a focus on phrases related to "Helvetia" or similar terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    HAEL
    -0.78
     Cunning
    -0.70
    rers
    -0.67
    æĦ
    -0.67
    CLASSIFIED
    -0.66
    EED
    -0.65
    parts
    -0.65
    EAR
    -0.65
    hyde
    -0.63
     Memories
    -0.63
    POSITIVE LOGITS
    idon
    0.94
    mut
    0.94
    abor
    0.92
    ios
    0.91
    pless
    0.91
    mand
    0.91
    ped
    0.85
    strings
    0.84
    mel
    0.84
    ping
    0.82
    Act Density 0.016%

    No Known Activations