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

    Neuronpedia

    INDEX
    Explanations

    phrases related to purpose or intention

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the purpose or consequence of actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ARK
    -0.65
    ohm
    -0.65
    hal
    -0.64
    ãĥ¡
    -0.63
    natureconservancy
    -0.62
    Drag
    -0.61
    notice
    -0.61
    AND
    -0.60
    DAQ
    -0.60
    cms
    -0.60
    POSITIVE LOGITS
    ndra
    0.80
    soever
    0.72
     they
    0.71
     whoever
    0.71
    atever
    0.67
     nobody
    0.66
     preserves
    0.65
     astronauts
    0.65
     borrowers
    0.64
     persisted
    0.64
    Act Density 0.077%

    No Known Activations