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

    Neuronpedia

    INDEX
    Explanations

    sequences of whitespace or tab characters in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    PART followed by text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kaldı
    -0.37
     témoin
    -0.34
     aanv
    -0.32
     confianza
    -0.31
    MockBean
    -0.30
    dAtA
    -0.30
     hediye
    -0.29
    Смотрите
    -0.29
     Shark
    -0.29
    berday
    -0.29
    POSITIVE LOGITS
    BagLayout
    0.72
    GenerationType
    0.60
    \{\\
    0.59
     InputDecoration
    0.54
     <<<<<<<<<<<<<<
    0.54
    XmlAccessType
    0.52
     BoxDecoration
    0.52
    évaluateur
    0.52
     Verſ
    0.51
    Tikang
    0.51
    Act Density 0.017%

    No Known Activations