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

    Neuronpedia

    INDEX
    Explanations

    various types of underscores and hyphens in the text

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

    URLs and emails

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    (
    -0.41
     aplicable
    -0.38
     want
    -0.38
     cár
    -0.36
    -
    -0.36
     receive
    -0.35
     récents
    -0.34
     comprob
    -0.33
     loopt
    -0.33
     decenas
    -0.33
    POSITIVE LOGITS
    ſehen
    0.90
    0.81
    niſſe
    0.79
     betweenstory
    0.79
    ſicht
    0.79
    +#+
    0.78
     propOrder
    0.78
    ⟬
    0.76
    #+#
    0.75
    ſſung
    0.75
    Act Density 0.023%

    No Known Activations