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

    Neuronpedia

    INDEX
    Explanations

    rights reserved, pun intended

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kendini
    0.43
    butanol
    0.43
     বলেছে
    0.41
    ಿಸಿಕೊಳ್ಳ
    0.40
     চা
    0.40
     commutator
    0.38
    Extraction
    0.38
     হার
    0.38
    extraction
    0.38
    aider
    0.38
    POSITIVE LOGITS
     reserved
    0.85
     Reserved
    0.78
    Reserved
    0.72
    reserved
    0.66
     réserv
    0.63
     reserving
    0.61
     riserv
    0.60
     rezerv
    0.59
     reserve
    0.58
     reserva
    0.57
    Act Density 0.000%

    No Known Activations