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

    Neuronpedia

    INDEX
    Explanations

    that matter

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     doct
    0.39
     sull
    0.35
     DO
    0.35
     st
    0.35
     enclosure
    0.34
     Une
    0.34
     stirred
    0.34
     ston
    0.33
     lö
    0.33
     ornaments
    0.33
    POSITIVE LOGITS
     verwenden
    0.45
    zb
    0.44
    <unused400>
    0.44
    <unused2042>
    0.44
    <unused745>
    0.43
     নির্ণ
    0.43
    <unused2187>
    0.42
    <unused2064>
    0.42
    <unused2132>
    0.42
    ց
    0.42
    Act Density 0.001%

    No Known Activations