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

    Neuronpedia

    INDEX
    Explanations

    directly impacted or affected

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    established
    0.63
    deterministic
    0.61
     Edin
    0.60
     established
    0.59
    Rivers
    0.58
    身長
    0.58
    solve
    0.58
    solving
    0.58
     Chis
    0.57
     solve
    0.57
    POSITIVE LOGITS
     affected
    1.65
     Affected
    1.55
     victims
    1.39
     impacted
    1.37
     victim
    1.35
    Affected
    1.34
     suffer
    1.32
     vulnerable
    1.32
     suffers
    1.31
     betroffen
    1.30
    Act Density 0.691%

    No Known Activations