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

    Neuronpedia

    INDEX
    Explanations

    LLM

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ль
    1.36
    esquerda
    1.34
    ு
    1.27
    ר
    1.27
    س
    1.25
    רת
    1.24
    P
    1.21
    רק
    1.19
    S
    1.18
    ва
    1.16
    POSITIVE LOGITS
     Dutta
    1.25
     westbound
    1.09
    ৮
    1.07
     brunt
    1.04
     lateribus
    1.02
     accessori
    1.02
     stare
    1.01
    将于
    1.00
    ুদ্ধে
    0.98
    说
    0.95
    Act Density 0.055%

    No Known Activations