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

    Neuronpedia

    INDEX
    Explanations

    safety guidelines and human reviewers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     padi
    0.45
    itriangular
    0.42
     nautical
    0.41
     rigidbody
    0.41
     intric
    0.40
     dibujo
    0.40
     wristwatch
    0.40
     aquare
    0.40
     trypsin
    0.39
     besi
    0.39
    POSITIVE LOGITS
     censorship
    0.64
     OpenAI
    0.60
     politič
    0.60
     cybersecurity
    0.59
     Metaverse
    0.58
     TikTok
    0.57
     disinformation
    0.56
     metaverse
    0.55
    ユーザー
    0.54
     users
    0.54
    Act Density 0.862%

    No Known Activations