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

    Neuronpedia

    INDEX
    Explanations

    instances of specific punctuation and conjunctions in a context that suggests dialogue or lists

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uber
    -0.16
    pper
    -0.15
    baugh
    -0.15
    ugh
    -0.15
    ì§ľ
    -0.15
    ter
    -0.14
    dash
    -0.14
    ughs
    -0.14
    uye
    -0.14
    UTERS
    -0.13
    POSITIVE LOGITS
     feeling
    0.22
     having
    0.22
     after
    0.21
     faced
    0.21
     ever
    0.21
     Having
    0.20
     knowing
    0.20
     through
    0.19
     flush
    0.19
     via
    0.19
    Act Density 0.092%

    No Known Activations