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

    Neuronpedia

    INDEX
    Explanations

    technical explanations or instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cycl
    0.49
     rétro
    0.49
     grinned
    0.48
     perturbed
    0.47
     Moroccan
    0.46
     retro
    0.46
     Sax
    0.45
     CID
    0.45
     COULD
    0.45
    сів
    0.45
    POSITIVE LOGITS
    म
    0.61
    B
    0.55
    m
    0.51
    /*
    0.49
    $$\
    0.49
    '''
    0.49
    U
    0.49
    R
    0.47
    F
    0.47
    #'
    0.45
    Act Density 0.001%

    No Known Activations