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

    Neuronpedia

    INDEX
    Explanations

    clauses introducing contrast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    שא
    0.52
    薷
    0.50
    ס
    0.49
    eresa
    0.48
    ENSION
    0.48
    ומים
    0.47
    ותו
    0.47
    ד
    0.47
     cháu
    0.46
    Subscription
    0.45
    POSITIVE LOGITS
     tinge
    0.47
     decoy
    0.47
     sites
    0.46
     roadblock
    0.46
     cross
    0.46
     mass
    0.45
    yni
    0.45
     lead
    0.45
    ;
    0.44
     فقط
    0.44
    Act Density 0.003%

    No Known Activations