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

    Neuronpedia

    INDEX
    Explanations

    references to direction and movement

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

    a heading or path

    np_acts-logits-general · gemini-2.0-flash

    direction and its modifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    stufen
    -0.55
     Stufe
    -0.47
    baden
    -0.46
    чество
    -0.45
    Act
    -0.45
    fiés
    -0.45
    ;#
    -0.43
     pozn
    -0.41
    ictwa
    -0.41
    Excluded
    -0.41
    POSITIVE LOGITS
     direction
    3.16
     directions
    2.66
    direction
    2.54
     Direction
    2.50
     DIRECTION
    2.35
    Direction
    2.23
    directions
    2.11
     Directions
    2.04
     dirección
    1.96
    方向
    1.89
    Act Density 0.259%

    No Known Activations