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

    Neuronpedia

    INDEX
    Explanations

    seeing things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )،
    0.92
    ”،
    0.91
    "،
    0.88
     Describes
    0.86
    ()=>{
    0.85
    »،
    0.84
     ځای
    0.84
     analys
    0.84
    fte
    0.81
    CFT
    0.80
    POSITIVE LOGITS
    ਾ
    1.12
    ת
    1.04
    ING
    1.02
    म
    1.02
    는
    0.99
    <0x98>
    0.98
    ла
    0.96
    ą
    0.88
    ية
    0.84
    נ
    0.84
    Act Density 0.267%

    No Known Activations