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

    Neuronpedia

    INDEX
    Explanations

    specific keywords or phrases from various texts, such as names ('Ross'), terms related to creation ('Created', 'Release'), and informational cues ('Step', 'Password')

    oai_token-act-pair · gpt-3.5-turbo

    terms and phrases related to software, documentation, and user instructions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     latter
    -0.64
     mush
    -0.60
     thereto
    -0.59
     dri
    -0.57
     behav
    -0.56
     cataly
    -0.56
     comparatively
    -0.56
     otherwise
    -0.55
     fertile
    -0.54
     poles
    -0.54
    POSITIVE LOGITS
    resa
    1.26
    odore
    1.17
    anmar
    1.01
    agascar
    0.88
    romeda
    0.86
    xiety
    0.83
    foundland
    0.82
    withstanding
    0.82
    mosp
    0.81
    jamin
    0.80
    Act Density 0.794%

    No Known Activations