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

    Neuronpedia

    INDEX
    Explanations

    text related to actions involving daring, risk-taking, and potentially controversial behavior

    oai_token-act-pair · gpt-3.5-turbo

    phrases associated with negative actions or comments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ndra
    -0.79
    ntil
    -0.77
    tions
    -0.74
    lich
    -0.70
    -+-+
    -0.69
    ambo
    -0.68
    etheless
    -0.68
    lished
    -0.65
    ategories
    -0.63
    vertisement
    -0.63
    POSITIVE LOGITS
     aback
    1.09
     seriously
    1.07
     cues
    1.03
     liberties
    1.02
     stride
    0.91
     Seriously
    0.90
     hostage
    0.90
     cue
    0.89
     plunge
    0.89
     precautions
    0.89
    Act Density 0.389%

    No Known Activations