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

    Neuronpedia

    INDEX
    Explanations

    phrases related to movement or changes in direction

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

    winding up or down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kuma
    -0.34
     asumir
    -0.34
     dziecko
    -0.32
    🏾
    -0.32
     fijar
    -0.32
     leps
    -0.32
    かります
    -0.32
    DEBUG
    -0.32
     granic
    -0.32
     República
    -0.31
    POSITIVE LOGITS
     Winding
    1.02
     winding
    0.99
    winding
    0.89
     windings
    0.83
     unwind
    0.74
     wound
    0.74
     Winder
    0.74
    wound
    0.73
     Wind
    0.72
     wind
    0.70
    Act Density 0.006%

    No Known Activations