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

    Neuronpedia

    INDEX
    Explanations

    empower, redefine, innovate

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     goodies
    0.55
     relativamente
    0.55
     horrendous
    0.55
    悲
    0.55
     wacky
    0.53
     deleterious
    0.52
     putative
    0.51
    そのような
    0.50
    基本的な
    0.50
     sobbing
    0.49
    POSITIVE LOGITS
     smarter
    0.68
     redefine
    0.66
     redefining
    0.65
     reinvent
    0.62
     reimag
    0.62
     Momentum
    0.60
     Empower
    0.59
     empowered
    0.59
     empower
    0.58
     uncompromising
    0.57
    Act Density 0.086%

    No Known Activations