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

    Neuronpedia

    INDEX
    Explanations

    yes/no answers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     culprit
    0.75
     upright
    0.70
    esteps
    0.68
    sière
    0.68
     Penguin
    0.66
    доби
    0.65
     true
    0.64
     Luckily
    0.64
     claws
    0.64
     similarities
    0.62
    POSITIVE LOGITS
    No
    1.09
    no
    1.06
     পণ্য
    0.93
    NO
    0.92
    娛樂
    0.88
    ／
    0.86
     No
    0.85
    இய
    0.84
    军事
    0.84
     או
    0.84
    Act Density 0.050%

    No Known Activations