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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks and their relation to content and emotional undertones

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

    instead or contrasting ideas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ftagPool
    -0.43
    évaluateur
    -0.41
    хьтан
    -0.40
    <bos>
    -0.40
    Tro
    -0.40
    المناصب
    -0.39
     intptr
    -0.38
    tvguidetime
    -0.37
    adan
    -0.35
    Datuak
    -0.35
    POSITIVE LOGITS
     instead
    0.54
     vielmehr
    0.51
    instead
    0.51
    反而
    0.51
    Instead
    0.50
    AnchorTagHelper
    0.49
    Referenties
    0.47
    batore
    0.47
     Instead
    0.46
    MessageOf
    0.46
    Act Density 0.072%

    No Known Activations