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

    Neuronpedia

    INDEX
    Explanations

    matlab code descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Borges
    0.46
     Bond
    0.42
     стиле
    0.42
     খ্যা
    0.41
    ंजे
    0.40
     ലോക
    0.40
     торго
    0.38
     Mulligan
    0.38
    癃
    0.38
     Compl
    0.37
    POSITIVE LOGITS
     ReLU
    0.58
     %
    0.53
     arxiv
    0.50
    %
    0.49
     %),
    0.49
     spectrogram
    0.49
    %\
    0.46
    %%%%%%%%%%%%%%%%
    0.46
     astrocyte
    0.46
     isotropic
    0.46
    Act Density 0.000%

    No Known Activations