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

    Neuronpedia

    INDEX
    Explanations

    positive evaluation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     HRM
    0.25
    MAS
    0.24
     ligger
    0.24
    copies
    0.24
    marginLeft
    0.24
     अजी
    0.23
     Copies
    0.23
    ობს
    0.22
    GFR
    0.22
    ?<
    0.22
    POSITIVE LOGITS
     quando
    0.30
     যখন
    0.29
     когда
    0.27
     when
    0.26
     عندما
    0.25
     cuando
    0.25
     случае
    0.25
     когато
    0.24
     whenever
    0.23
     situations
    0.23
    Act Density 0.018%

    No Known Activations