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

    Neuronpedia

    INDEX
    Explanations

    importance of tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     çeşitli
    0.55
     expts
    0.54
     лише
    0.53
    primarily
    0.49
    okay
    0.49
     Expt
    0.49
     soln
    0.48
    Incidentally
    0.47
     Admittedly
    0.46
    Owing
    0.45
    POSITIVE LOGITS
     wich
    1.14
    ´
    1.13
     frecu
    1.11
     inmedi
    1.08
     significative
    1.08
     proyect
    1.03
     adimensional
    1.02
     lenght
    1.00
     caracter
    0.98
     Departament
    0.97
    Act Density 0.117%

    No Known Activations