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

    Neuronpedia

    INDEX
    Explanations

    I + statement of sentiment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     isotonic
    0.43
     illumination
    0.43
     Wherever
    0.43
     bulb
    0.42
     phenomenon
    0.41
     segments
    0.41
     runtime
    0.41
     Phenomen
    0.41
     runes
    0.41
     Bulb
    0.40
    POSITIVE LOGITS
     reluctantly
    0.68
     urge
    0.62
     disgusted
    0.59
     abhor
    0.58
     sincerely
    0.57
     apologised
    0.57
     толькі
    0.57
     hoped
    0.57
     willing
    0.55
     earnestly
    0.55
    Act Density 0.002%

    No Known Activations