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

    Neuronpedia

    INDEX
    Explanations

    instruction patterns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    .
    0.37
    સ
    0.33
    ak
    0.31
    к
    0.29
    ं
    0.29
    ad
    0.29
    ﻧ
    0.29
    ان
    0.28
    い
    0.28
    ب
    0.28
    POSITIVE LOGITS
     Gosudarstvennyj
    0.31
    thisComponent
    0.29
    IOR
    0.28
    issanti
    0.27
     patitth
    0.26
    naires
    0.26
    DataDiv
    0.26
     தடு
    0.26
    बंधनाच्या
    0.25
     Influenza
    0.25
    Act Density 0.001%

    No Known Activations