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

    Neuronpedia

    INDEX
    Explanations

    code structures

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     hoes
    -0.09
     illuminated
    -0.08
     hemp
    -0.08
     noire
    -0.07
     মহান
    -0.07
     masking
    -0.07
     heroic
    -0.07
    /site
    -0.07
     illumination
    -0.07
    य
    -0.07
    POSITIVE LOGITS
     Nested
    0.11
    Nested
    0.10
     nested
    0.10
     wiederum
    0.10
    nested
    0.10
    Nest
    0.09
     отдельно
    0.09
     Binnen
    0.08
    里面
    0.08
     влож
    0.08
    Act Density 0.006%

    No Known Activations