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

    Neuronpedia

    INDEX
    Explanations

    create files and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     unsuspecting
    0.58
     gleaned
    0.52
     unmet
    0.52
     unused
    0.51
     discomfort
    0.50
     uncomfortable
    0.48
     unhappy
    0.48
     remnant
    0.48
     stolen
    0.48
     inaction
    0.48
    POSITIVE LOGITS
     এবং
    0.59
     আৰু
    0.59
    Secondo
    0.58
     మరియు
    0.58
    מו
    0.58
     Examinations
    0.58
    必要的
    0.57
     ferries
    0.57
     необходимых
    0.57
    Guides
    0.56
    Act Density 0.296%

    No Known Activations