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

    Neuronpedia

    INDEX
    Explanations

    Single items/details

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     schafft
    -0.08
    راک
    -0.08
     übernimmt
    -0.08
    modifier
    -0.08
     intervient
    -0.08
     Ribeiro
    -0.08
    stemming
    -0.08
     Pitt
    -0.08
     innebär
    -0.08
     aporta
    -0.08
    POSITIVE LOGITS
     strangely
    0.10
     handwriting
    0.10
     corrupted
    0.09
     scrib
    0.09
     inexp
    0.09
     astonishing
    0.09
     surprisingly
    0.09
     seemingly
    0.09
     handwritten
    0.08
     bizarre
    0.08
    Act Density 0.061%

    No Known Activations