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

    Neuronpedia

    INDEX
    Explanations

    references to images and visual data formats

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

    ![ figures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    transQ
    -0.86
    OGND
    -0.84
    AddTagHelper
    -0.77
    :+:
    -0.75
    <pad>
    -0.74
    <unused14>
    -0.74
    <unused8>
    -0.73
    <unused28>
    -0.73
    <unused3>
    -0.73
    [@BOS@]
    -0.73
    POSITIVE LOGITS
     freien
    0.35
     Schlacht
    0.33
     Bélgica
    0.33
     selben
    0.33
    Player
    0.30
     tatuajes
    0.30
    F
    0.29
     Crespo
    0.29
     miteinander
    0.29
     schweren
    0.29
    Act Density 0.003%

    No Known Activations