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

    Neuronpedia

    INDEX
    Explanations

    run script

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    带领
    0.66
     Kauff
    0.60
     Muda
    0.56
     systemat
    0.55
    MDg
    0.54
     Beyer
    0.54
    arantad
    0.54
    巽
    0.52
     jan
    0.51
    あっ
    0.50
    POSITIVE LOGITS
    direct
    0.60
    Direct
    0.58
     Direct
    0.57
     direct
    0.54
     직접
    0.54
    directly
    0.53
     scripts
    0.53
     script
    0.52
    uvian
    0.51
    loader
    0.50
    Act Density 0.060%

    No Known Activations