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

    Neuronpedia

    INDEX
    Explanations

    detective, deduction, investigation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ensl
    0.52
     alegria
    0.48
     Powerful
    0.43
     powering
    0.43
     ofens
    0.43
    estinal
    0.42
     यज्ञ
    0.42
     Nge
    0.41
     ጎ
    0.41
     Byte
    0.41
    POSITIVE LOGITS
     detective
    1.12
    Detective
    0.95
     detectives
    0.94
     Detective
    0.94
     Sherlock
    0.91
    推理
    0.88
    🕵
    0.84
     kriminal
    0.80
     crime
    0.77
     sle
    0.77
    Act Density 0.216%

    No Known Activations