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

    Neuronpedia

    INDEX
    Explanations

    phrases related to initiating new actions or projects

    oai_token-act-pair · gpt-3.5-turbo

    actions related to opening or facilitating processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Niet
    -0.64
     anecd
    -0.61
    ãĤ©
    -0.55
    renheit
    -0.54
     undermin
    -0.52
     terribly
    -0.51
     preceded
    -0.51
     nostalg
    -0.51
     Vaugh
    -0.51
    stood
    -0.51
    POSITIVE LOGITS
     RELEASE
    0.70
     ASAP
    0.68
    ¶
    0.60
     Xperia
    0.58
    )?
    0.57
     Belfast
    0.56
    ↵
    0.54
    "?
    0.54
    '?
    0.53
     Image
    0.52
    Act Density 0.968%

    No Known Activations