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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences instructing to take a specific action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "leave" and its variations used in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    alist
    -0.83
    kered
    -0.74
     Cosponsors
    -0.69
    insula
    -0.68
    reme
    -0.67
    gio
    -0.67
    Eva
    -0.65
    reader
    -0.64
    andan
    -0.63
    DX
    -0.61
    POSITIVE LOGITS
     undone
    0.92
    overs
    0.90
     aside
    0.76
     untreated
    0.75
    wich
    0.74
     unfinished
    0.73
    ipp
    0.70
     Behind
    0.69
    ings
    0.68
     behind
    0.67
    Act Density 0.039%

    No Known Activations