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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "when" indicating conditional or temporal situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    isi
    -0.17
    abd
    -0.17
    oley
    -0.15
    alis
    -0.15
     abund
    -0.14
    rement
    -0.13
     pall
    -0.13
     redistributed
    -0.13
    abstract
    -0.13
    isa
    -0.13
    POSITIVE LOGITS
    ãĥ¼ãĥģ
    0.15
    woord
    0.15
    artz
    0.15
     followed
    0.14
    ázd
    0.14
    gli
    0.14
    oundingBox
    0.14
    ucket
    0.14
    manual
    0.14
    ampire
    0.14
    Act Density 0.068%

    No Known Activations