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

    Neuronpedia

    INDEX
    Explanations

    Scorsese directs, follower, understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     suggest
    0.51
     recommend
    0.49
     terminals
    0.47
     wines
    0.46
     doigts
    0.46
     backpacks
    0.46
     IEC
    0.45
     is
    0.45
     Recommend
    0.44
     recommends
    0.44
    POSITIVE LOGITS
    ሌ
    0.52
    ارين
    0.50
    r
    0.49
    ")
    0.48
    डिंग
    0.48
    تون
    0.47
     ষে
    0.46
    ﺱ
    0.46
    ræ
    0.45
    mama
    0.44
    Act Density 0.000%

    No Known Activations