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

    Neuronpedia

    INDEX
    Explanations

    shit and vulgarity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    बिर
    0.44
     lộ
    0.42
    ∕
    0.40
    icuously
    0.39
     continueRoutine
    0.38
    দ্ম
    0.38
    ❙
    0.38
    inFile
    0.37
     大き
    0.36
     तिर
    0.36
    POSITIVE LOGITS
     shit
    3.64
     crap
    3.19
     Shit
    3.09
    shit
    2.92
     bullshit
    2.20
     shitty
    1.95
     merda
    1.76
     fucked
    1.57
    💩
    1.55
     crappy
    1.53
    Act Density 0.022%

    No Known Activations