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

    Neuronpedia

    INDEX
    Explanations

    carries emotional, tell my, before errors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     मंच
    0.39
    ("|
    0.37
    ֿ
    0.36
     დიდ
    0.36
     Boż
    0.35
     endeavors
    0.35
    omass
    0.34
     swój
    0.34
     passionately
    0.34
     Naz
    0.33
    POSITIVE LOGITS
    locks
    0.41
    ges
    0.38
    ண்டும்
    0.38
    falt
    0.38
     आटा
    0.36
    lera
    0.36
     halve
    0.36
    gre
    0.36
    leri
    0.35
    fa
    0.35
    Act Density 0.000%

    No Known Activations