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

    Neuronpedia

    INDEX
    Explanations

    things that are tricky

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     unimaginable
    0.64
     enormously
    0.57
     огромное
    0.57
     inconceivable
    0.54
     огром
    0.52
     невероят
    0.51
     enormes
    0.50
     enormous
    0.50
     niezwy
    0.50
     inconce
    0.49
    POSITIVE LOGITS
     tricky
    1.77
     challenging
    0.99
     iff
    0.91
     problematic
    0.89
     treacherous
    0.89
     dodgy
    0.89
     risky
    0.88
     tough
    0.86
     bumpy
    0.86
    challeng
    0.86
    Act Density 0.279%

    No Known Activations