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

    Neuronpedia

    INDEX
    Explanations

    HTML attributes and values in code

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

    inner div

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <bos>
    -0.40
     gebeten
    -0.34
     kijang
    -0.28
    localctx
    -0.28
    Viki
    -0.27
    veien
    -0.26
     wijk
    -0.26
    Groetjes
    -0.25
    gól
    -0.25
     vapa
    -0.25
    POSITIVE LOGITS
     Hidden
    0.90
    hidden
    0.88
     hidden
    0.86
    Hidden
    0.86
    HIDDEN
    0.75
     surla
    0.73
    tagHelperRunner
    0.70
    fjspx
    0.68
     concealed
    0.68
    ruptedException
    0.65
    Act Density 0.001%

    No Known Activations