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

    Neuronpedia

    INDEX
    Explanations

    contrasting ideas or counterpoint

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     specifics
    0.44
    僞
    0.44
     eyepiece
    0.42
     obeyed
    0.42
     specifically
    0.41
     fø
    0.41
     glossy
    0.41
     zwar
    0.41
     bindings
    0.41
     mx
    0.41
    POSITIVE LOGITS
     reconsider
    0.54
     sollten
    0.49
     reconsideration
    0.47
    should
    0.46
     overlooked
    0.46
     debería
    0.45
     deveria
    0.45
     devrait
    0.45
    越来越多的
    0.45
     overlooking
    0.44
    Act Density 0.045%

    No Known Activations