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

    Neuronpedia

    INDEX
    Explanations

    shapes of letters

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     reward
    -0.09
     વ્યક્ત
    -0.08
    reward
    -0.08
    -debug
    -0.08
     ($_
    -0.08
    Encode
    -0.08
     journ
    -0.08
     unwind
    -0.08
     અથવા
    -0.08
     conversations
    -0.08
    POSITIVE LOGITS
     shaft
    0.09
     piercing
    0.09
     supl
    0.09
     épa
    0.08
     shafts
    0.08
     haar
    0.08
     punched
    0.08
     sleeves
    0.08
     delantero
    0.08
     scoop
    0.08
    Act Density 0.005%

    No Known Activations