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

    Neuronpedia

    INDEX
    Explanations

    Generating access to content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    bor
    0.46
    Islamic
    0.44
    prom
    0.43
    Kansas
    0.41
    dif
    0.39
    Diploma
    0.39
    Pakistan
    0.38
    Pacific
    0.38
    mer
    0.38
    personal
    0.38
    POSITIVE LOGITS
    窅
    0.47
     peuvent
    0.46
     semblent
    0.46
     podrían
    0.45
     finne
    0.43
     pueden
    0.43
     attenu
    0.43
    の原因
    0.43
    覚
    0.43
    唃
    0.43
    Act Density 0.000%

    No Known Activations