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

    Neuronpedia

    INDEX
    Explanations

    names preceded by "ini" as a suffix

    oai_token-act-pair · gpt-3.5-turbo

    mention of specific individuals or names

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    manship
    -0.82
     mathemat
    -0.76
    spin
    -0.76
    deck
    -0.75
    boards
    -0.73
    lisher
    -0.72
    lehem
    -0.71
    taining
    -0.71
    theless
    -0.70
    saw
    -0.69
    POSITIVE LOGITS
    Äĩ
    1.03
    emi
    0.97
    zzle
    0.97
    zzo
    0.96
    acs
    0.94
    otti
    0.93
    ya
    0.92
    ère
    0.87
    oti
    0.85
    opsis
    0.84
    Act Density 0.020%

    No Known Activations