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

    Neuronpedia

    INDEX
    Explanations

    references to counting or numerical data

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

    counting things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Administrativna
    -0.71
    +#+#
    -0.68
     <=",
    -0.48
    /**
    -0.48
    AnimationsModule
    -0.47
    ackerel
    -0.46
    ----</
    -0.46
     Italijani
    -0.45
     дописавши
    -0.45
     GenerationType
    -0.45
    POSITIVE LOGITS
     counting
    1.24
     counted
    1.23
     count
    1.21
     counts
    1.16
     Count
    1.12
     COUNT
    1.10
     Counting
    1.09
    counted
    1.07
    counting
    1.05
    count
    1.03
    Act Density 0.026%

    No Known Activations