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

    Neuronpedia

    INDEX
    Explanations

    cross-lingual conjunctions and connectors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     have
    0.48
     varieties
    0.41
     B
    0.40
     bescherm
    0.38
     –
    0.38
     protection
    0.38
     care
    0.38
     MPs
    0.38
     Festivals
    0.38
     foundational
    0.37
    POSITIVE LOGITS
    ानुसार
    0.53
    暠
    0.51
    এবং
    0.50
    blurred
    0.47
    이며
    0.46
    𝘫
    0.46
     тихо
    0.46
     अा
    0.46
    하려고
    0.45
    และ
    0.45
    Act Density 0.002%

    No Known Activations