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

    Neuronpedia

    INDEX
    Explanations

    words and phrases related to safety, regulations, and procedures

    oai_token-act-pair · gpt-3.5-turbo

    aspects related to regulations and safety concerns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hypocr
    -0.58
    ................................................................
    -0.56
     Insp
    -0.53
     pengu
    -0.53
     Rah
    -0.53
    omsky
    -0.53
     Pocket
    -0.52
    anan
    -0.52
     Nath
    -0.52
     condem
    -0.51
    POSITIVE LOGITS
    ().
    0.67
    etheless
    0.64
     steroids
    0.62
    ween
    0.61
    req
    0.60
    NetMessage
    0.60
    ccess
    0.59
    )).
    0.57
    tnc
    0.57
     msec
    0.57
    Act Density 1.141%

    No Known Activations