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

    Neuronpedia

    INDEX
    Explanations

    phrases prompting to check out something like articles or videos

    oai_token-act-pair · gpt-3.5-turbo

    phrases that encourage checking out or exploring additional content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enei
    -0.73
     dictate
    -0.71
     awake
    -0.66
     fabrication
    -0.65
    ajor
    -0.64
     rightful
    -0.63
    ylum
    -0.63
     conviction
    -0.63
     grasped
    -0.63
    Fourth
    -0.62
    POSITIVE LOGITS
     whats
    0.81
    nels
    0.68
    casts
    0.67
     how
    0.65
    icles
    0.65
    posts
    0.63
     www
    0.62
    flows
    0.62
     Braun
    0.61
    fitted
    0.61
    Act Density 0.024%

    No Known Activations