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

    Neuronpedia

    INDEX
    Explanations

    references to data extraction or sourcing in a scientific context

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

    extracted from

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     faſt
    -0.47
     Monfieur
    -0.46
     skydd
    -0.42
     igång
    -0.40
     tarko
    -0.40
     puol
    -0.38
     médaille
    -0.38
     perfección
    -0.38
     sää
    -0.38
     varandra
    -0.38
    POSITIVE LOGITS
    extracted
    0.67
     extracted
    0.60
    taken
    0.59
     StringTokenizer
    0.58
     gleaned
    0.57
    翻攝
    0.56
     taken
    0.54
    pulled
    0.54
    Stolen
    0.54
     Taken
    0.54
    Act Density 0.060%

    No Known Activations