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

    Neuronpedia

    INDEX
    Explanations

    always

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ereo
    -0.08
     उप
    -0.07
     perfor
    -0.07
     bif
    -0.06
     Petro
    -0.06
    57
    -0.06
    'post
    -0.06
    58
    -0.06
     homicide
    -0.06
     retrofit
    -0.06
    POSITIVE LOGITS
     always
    0.13
     Always
    0.12
    always
    0.11
    Always
    0.11
     ALWAYS
    0.10
    Definitions
    0.08
    Last
    0.08
     Lars
    0.08
     ever
    0.08
    _ALWAYS
    0.08
    Act Density 0.033%

    No Known Activations