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

    Neuronpedia

    INDEX
    Explanations

    important notes and disclaimers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    là
    0.59
    󰀄
    0.58
     самым
    0.55
    Là
    0.54
    手里
    0.53
    嚷
    0.53
     самая
    0.53
    Dangerous
    0.53
    dangerous
    0.52
     বিনি
    0.52
    POSITIVE LOGITS
     note
    1.55
     Note
    1.54
    Note
    1.35
    note
    1.19
     NOTE
    1.15
     noted
    1.01
     notes
    1.00
    NOTE
    0.96
     Notes
    0.91
     noting
    0.90
    Act Density 0.091%

    No Known Activations